papers

# Auton Lab Publications

## A Bayesian Spatial Scan Statistic

Authors: Daniel Neill, Andrew Moore, Gregory Cooper

This paper develops a new Bayesian method for cluster detection, the “Bayesian spatial scan statistic,” and compares this method to the standard (frequentist) scan statistic approach on the task of prospective disease surveillance.

Authors: Anna Goldenberg, Jeremy Kubica, Paul Komarek, Andrew Moore, Jeff Schneider

Link data, consisting of a collection of subsets of entities, can be an important source of information for a variety of fields including the social sciences, biology, criminology, and business intelligence. However, these links may be incomplete, containing one or more unknown members. We consider the problem of link completion, identifying which entities are the most likely missing members of a link given the previously observed links. We concentrate on the case of one missing entity. We compare a variety of recently developed along with standard machine learning and strawman algorithms adjusted to suit the task. The algorithms were tested extensively on a simulated and a range of real-world data sets.

## A Composite Likelihood View For Multi-Label Classification

Given limited training samples, learning to classify multiple labels is challenging. Problem decomposition is widely used in this case, where the original problem is decomposed into a set of easier-to-learn subproblems, and predictions from subproblems are combined to make the final decision.

In this paper we show the connection between composite likelihoods and many multilabel decomposition methods, e.g., one-vs-all, one-vs-one, calibrated label ranking, probabilistic classifier chain. This connection holds promise for improving problem decomposition in both the choice of subproblems and the combination of subproblem decisions.

As an attempt to exploit this connection, we design a composite marginal method that improves pairwise decomposition. Pairwise label comparisons, which seem to be a natural choice for subproblems, are replaced by bivariate label densities, which are more informative and natural components in a composite likelihood. For combining subproblem de- cisions, we propose a new mean-field approximation that minimizes the notion of composite divergence and is potentially more robust to inaccurate estimations in subproblems.

Empirical studies on five data sets show that, given limited training samples, the proposed method outperforms many alternatives.

Authors: Yi Zhang and Jeff Schneider

## A Constraint Generation Approach to Learning Stable Linear Dynamical Systems

Authors: Sajid Siddiqi, Byron Boots, Geoff Gordon

Stability is a desirable characteristic for linear dynamical systems, but it is often ignored by algorithms that learn these systems from data. We propose a novel method for learning stable linear dynamical systems: we formulate an approximation of the problem as a convex program, start with a solution to a relaxed version of the program, and incrementally add constraints to improve stability. Rather than continuing to generate constraints until we reach a feasible solution, we test stability at each step; because the convex program is only an approximation of the desired problem, this early stopping rule can yield a higher-quality solution. We apply our algorithm to the task of learning dynamic textures from image sequences as well as to modeling biosurveillance drug-sales data. The constraint generation approach leads to noticeable improvement in the quality of simulated sequences. We compare our method to those of Lacy and Bernstein, with positive results in terms of accuracy, quality of simulated sequences, and efficiency.

## A Dynamic Adaptation of AD-Trees for Efficient Machine Learning on Large Data Sets

Authors: Paul Komarek, Andrew Moore

This talk was presented at ICML 2000. It describes a modification to AD-trees to allow incremental and lazy growth. We discuss our implementation of these Dynamic AD-trees and present results for datasets with scores of high-arity attributes and millions of rows. ICML 2000.

These slides are best understood with the help of my notes from the presentation.

## A Fast Multi-Resolution Method For Detection of Significant Spatial Disease Clusters

Authors: Daniel Neill, Andrew Moore

Given an N x N grid of squares, where each square has a count and an underlying population, our goal is to find the square region with the highest density, and to calculate its significance by randomization. Any density measure D, dependent on the total count and total population of a region, can be used. For example, if each count represents the number of disease cases occurring in that square, we can use Kulldorff's spatial scan statistic D_K to find the most significant spatial disease cluster. A naive approach to finding the maximum density region requires O(N^3) time, and is generally computationally infeasible. We present a novel algorithm which partitions the grid into overlapping regions, bounds the maximum score of subregions contained in each region, and prunes regions which cannot contain the maximum density region. For sufficiently dense regions, this method finds the maximum density region in optimal O(N^2) time, in practice resulting in significant (10-200x) speedups.

## A Fast Multi-Resolution Method for Detection of Significant Spatial Overdensities

Authors: Daniel Neill, Andrew Moore

Given an NxN grid of squares, where each square sij has count cij and an underlying population pij, our goal is to find the square region S with the highest density, and to calculate the significance of this region by Monte Carlo testing. Any density measure D, which depends on the total count and total population of the region, can be used. For example, if each count cij represents the number of disease cases occurring in that square, we can use Kulldorff's spatial scan statistic D_K to find the most significant spatial disease cluster. A naive approach to finding the region of maximum density would be to calculate the density measure for every square region: this requires O(RN^3) calculations, where R is the number of Monte Carlo replications, and hence is generally computationally infeasible. We present a novel multi-resolution algorithm which partitions the grid into overlapping regions, bounds the maximum score of subregions contained in each region, and prunes regions which cannot contain the maximum density region. For sufficiently dense regions, this method finds the maximum density region in optimal O(RN^2) time, and in practice it results in significant (10-200x) speedups as compared to the naive approach.

## A Latent Space Approach to Dynamic Embedding of Co-Occurrence Data

Authors: Purnamrita Sarkar, Sajid Siddiqi, Geoff Gordon

We consider dynamic co-occurrence data, such as author-word links in papers published in successive years of the same conference. For static co-occurrence data, researchers often seek an embedding of the entities (authors and words) into a low-dimensional Euclidean space. We generalize a recent static co-occurrence model, the CODE model of Globerson et al. (2004), to the dynamic setting: we seek coordinates for each entity at each time step. The coordinates can change with time to explain new observations, but since large changes are improbable, we can exploit data at previous and subsequent steps to find a better explanation for current observations. To make inference tractable, we show how to approximate our observation model with a Gaussian distribution, allowing the use of a Kalman filter for tractable inference. The result is the first algorithm for dynamic embedding of co-occurrence data which provides distributional information for its coordinate estimates. We demonstrate our model both on synthetic data and on author-word data from the NIPS corpus, showing that it produces intuitively reasonable embeddings. We also provide evidence for the usefulness of our model by its performance on an author-prediction task.

## A Machine Learning Approach for Dynamical Mass Measurements of Galaxy Clusters

Authors: Michelle Ntampaka, Hy Trac, Dougal J. Sutherland, Nicholas Battaglia, Barnabás Póczos, and Jeff Schneider

We present a modern machine learning approach for cluster dynamical mass measurements that is a factor of two improvement over using a conventional scaling relation. Different methods are tested against a mock cluster catalog constructed using halos with mass >= 10^14 Msolar/h from Multidark's publicly-available N-body MDPL halo catalog. In the conventional method, we use a standard M(sigmav) power law scaling relation to infer cluster mass, M, from line-of-sight (LOS) galaxy velocity dispersion, sigmav. The resulting fractional mass error distribution is broad, with width=0.87 (68% scatter), and has extended high-error tails. The standard scaling relation can be simply enhanced by including higher-order moments of the LOS velocity distribution. Applying the kurtosis as a correction term to log(sigma_v) reduces the width of the error distribution to 0.74 (16% improvement). Machine learning can be used to take full advantage of all the information in the velocity distribution. We employ the Support Distribution Machines (SDMs) algorithm that learns from distributions of data to predict single values. SDMs trained and tested on the distribution of LOS velocities yield width=0.46 (47% improvement). Furthermore, the problematic tails of the mass error distribution are effectively eliminated. Decreasing cluster mass errors will improve measurements of the growth of structure and lead to tighter constraints on cosmological parameters.

## A Nonparametric Approach to Noisy and Costly Optimization

Authors: Brigham Anderson, Andrew Moore, David Cohn

This paper describes Pairwise Bisection: a nonparametric approach to optimizing a noisy function with few function evaluations. The algorithm uses nonparametric resoning about simple geometric relationships to find minima efficiently. Two factors often frustrate optimization: noise and cost. Output can contain significant quantities of noise or error, while time or money allow for only a handful of experiments. Pairwise bisection is used here to attempt to automate the process of robust and efficient experiment design. Real world functions also tend to violate traditional assumptions of continuousness and Gaussian noise. Since nonparametric statistics do not depend on these assumptions, this algorithm can optimize a wide variety of phenomena with fewer restrictions placed on noise. The algorithm's performance is compared to that of three competing algorithms, Amoeba, PMAX, and Q2 on several different test functions. Results on these functions indicate competitive performance and superior resistance to noise.

## A study into detection of bio-events in multiple streams of surveillance data

Authors: Josep Roure, Artur Dubrawski, Jeff Schneider

This paper reviews the results of a study into combining evidence from multiple streams of surveillance data in order to improve timeliness and speciﬁcity of detection of bio-events. In the experiments we used three streams of real food- and agriculture-safety related data\ that is being routinely collected at slaughter houses across the nation, and which carry mutually complementary information about potential outbreaks of bio-events. The results indicate that: (1) Non-speciﬁc aggregation of p-values produced by event detectors set on individual streams of data can lead to superior detection power over that of the individual detectors, and (2) Design of multi-stream detectors tailored to the particular characteristics of the events of interest can further improve timeliness and speciﬁcity of detection. In a practical setup, we recommend combining a set of speciﬁc multi-stream detectors focused on individual types of predictable and deﬁnable scenarios of interest, with non-speciﬁc multi-stream detectors, to account for both anticipated and emerging types of bio-events.

## A tractable approach to finding closest truncated-commute-time neighbors in large graphs

Authors: Purnamrita Sarkar, Andrew W. Moore

Recently there has been much interest in graph-based learning, with applications in collaborative fltering for recommender networks, link prediction for social networks and fraud detection. These networks can consist of millions of entities, and so it is very important to develop highly effcient techniques. We are especially interested in accelerating random walk approaches to compute some very interesting proximity measures of these kinds of graphs. These measures have been shown to do well empirically (Liben-Nowell & Kleinberg, 2003; Brand, 2005). We introduce a truncated variation on a well-known measure, namely commute times arising from random walks on graphs. We present a very novel algorithm to compute all interesting pairs of approximate nearest neighbors in truncated commute times, without computing it between all pairs. We show results on both simulated and real graphs of size up to 100,000 entities, which indicate near-linear scaling in computation time.

## Active area search via Bayesean quadrature

Authors: Yifei Ma and Roman Garnett and Jeff Schneider

The selection of data collection locations is a problem that has received significant research attention from classical design of experiments to various recent active learning algorithms. Typical objectives are to map an unknown function, optimize it, or find level sets in it. Each of these objectives focuses on an assessment of individual points. The introduction of set kernels has led to algorithms that instead consider labels assigned to sets of data points. In this paper we combine these two concepts and consider the problem of choosing data collection locations when the goal is to identify regions whose set of collected data would be labeled positively by a set classifier. We present an algorithm for the case where the positive class is defined in terms of a region's average function value being above some threshold with high probability, a problem we call active area search. To this end, we model the latent function using a Gaussian process and use Bayesian quadrature to estimate its integral on predefined regions. Our method is the first which directly solves the active area search problem. In experiments it outperforms previous algorithms that were developed for other active search goals.

## Active learning and search on low-rank matrices

Authors: Dougal J. Sutherland, Barnabás Póczos, and Jeff Schneider

Collaborative prediction is a powerful technique, useful in domains from recommender systems to guiding the scientific discovery process. Low-rank matrix factorization is one of the most powerful tools for collaborative prediction. This work presents a general approach for active collaborative prediction with the Probabilistic Matrix Factorization model. Using variational approximations or Markov chain Monte Carlo sampling to estimate the posterior distribution over models, we can choose query points to maximize our understanding of the model, to best predict unknown elements of the data matrix, or to find as many “positive” data points as possible. We evaluate our methods on simulated data, and also show their applicability to movie ratings prediction and the discovery of drug-target interactions.

## Active Learning for Anomaly and Rare Category Detection

Authors: Dan Pelleg, Andrew Moore

We introduce a novel active-learning scenario in which a user wants to work with a learning algorithm to identify {em useful} anomalies. These are distinguished from the traditional statistical definition of anomalies as outliers or merely ill-modeled points. Our distinction is that the usefulness of anomalies is categorized subjectively by the user. We make two additional assumptions. First, there exist extremely few useful anomalies to be hunted down within a massive dataset. Second, both useful and useless anomalies may sometimes exist within tiny classes of similar anomalies. The challenge is thus to identify “rare category” records in an unlabeled noisy set with help (in the form of class labels) from a human expert who has a small budget of datapoints that they are prepared to categorize. We propose a technique to meet this challenge, which assumes a mixture model fit to the data, but otherwise makes no assumptions on the particular form of the mixture components. This property promises wide applicability in real-life scenarios and for various statistical models. We give an overview of several alternative methods, highlighting their strengths and weaknesses, and conclude with a detailed empirical analysis. We show that our method can quickly zoom in on an anomaly set containing a few tens of points in a dataset of hundreds of thousands.

## Active learning for identifying function threshold boundaries

Authors: Brent Bryan, Jeff Schneider, Robert C. Nichol, Christopher J. Miller, Christopher R. Genovese, Larry Wasserman

We present an efficient algorithm to actively select queries for learning the boundaries separating a function domain into regions where the function is above and below a given threshold. We develop experiment selection methods based on entropy, misclassification rate, variance, and their combinations, and show how they perform on a number of data sets. We then show how these algorithms are used to determine simultaneously valid $1-\alpha$ confidence intervals for seven cosmological parameters. Experimentation shows that the algorithm reduces the computation necessary for the parameter estimation problem by an order of magnitude.

## Discovery of Complex Anomalous Patterns of Sexual Violence in El Salvador

Authors: ​Maria De-Arteaga, Artur Dubrawski

Data for Policy, 2016.

Authors: William Herlands, Maria De-­Arteaga, Daniel Neill, and Artur Dubrawski

NIPS Workshop: Optimization for Machine Learning, 2015.

## Canonical Autocorrelation Analysis for Radiation Threat Detection

Authors: Maria De­-Arteaga, Artur Dubrawski, Peter Huggins

Heinz First Paper, Heinz College / Data Analysis Project, Machine Learning Department, 2016. Carnegie Mellon University.

## BRAINZOOM: High Resolution Reconstruction from Multi-modal Brain Signals

Authors: X. Fu and K. Huang and O. Stretcu and H. Song and E. Papalexakis and P. Talukdar and T. Mitchell and N. Sidiropoulos and C. Faloutsos and B. Poczos

Year: 2017

Venue: SIAM Data Mining

abstract unavailable at this time

## Query Efficient Posterior Estimation in Scientific Experiments via Bayesian Active Learning

Authors: K. Kandasamy and J. Schneider and B. Poczos

Year: 2016

Venue: Artificial Intelligence Journal

abstract unavailable at this time

## Enabling Dark Energy Science with Deep Generative Models of Galaxy Images

Authors: M. Ravanbakhsh and F. Lanusse and R. Mandelbaum and J. Schneider and B. Poczos

Year: 2017

Venue: Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17)

abstract unavailable at this time

## Quantifying Differences and Similarities in Whole-Brain White Matter Architecture Using Local Connectome Fingerprints

Authors: F. Yeh and J. Vettel and A. Singh and B. Poczos and S. Grafton and K. Erickson and W. Tseng and T. Verstynen

Year: 2016

Venue: PLOS Computational Biology

abstract unavailable at this time

## Learning Theory for Distribution Regression

Authors: Z. Szabo and B. Sriperumbudur and B. Poczos and A. Gretton

Year: 2016

Venue: Journal of Machine Learning Research

abstract unavailable at this time

## Gaussian Process Bandit Optimisation with Multi-fidelity Evaluations

Authors: K. Kandasamy and G. Dasarathy and J. Oliva and J. Schneider and B. Poczos

Year: 2016

Venue: Proceedings of the Neural Information Processing Systems (NIPS-16)

abstract unavailable at this time

## Efficient Nonparametric Smoothness Estimation

Authors: S. Singh and S. Du and B. Poczos

Year: 2016

Venue: Proceedings of the Neural Information Processing Systems (NIPS-16)

abstract unavailable at this time

## Fast Stochastic Methods for Nonsmooth Nonconvex Optimization

Authors: S. Reddi and S. Sra and B. Poczos and A. Smola

Year: 2016

Venue: Proceedings of the Neural Information Processing Systems (NIPS-16)

abstract unavailable at this time

## Variance Reduction in Stochastic Gradient Langevin Dynamics

Authors: A. Dubey and S. Reddi and S. Williamson and B. Poczos and A. Smola and E. Xing

Year: 2016

Venue: Proceedings of the Neural Information Processing Systems (NIPS-16)

abstract unavailable at this time

## Finite-Sample Analysis of Fixed-k Nearest Neighbor Density Functionals Estimators

Authors: S. Singh and B. Poczos

Year: 2016

Venue: Proceedings of the Neural Information Processing Systems (NIPS-16)

abstract unavailable at this time

## The Multi-fidelity Multi-armed Bandit

Authors: K. Kandasamy and G. Dasarathy and B. Poczos and J. Schneider

Year: 2016

Venue: Proceedings of the Neural Information Processing Systems (NIPS-16)

abstract unavailable at this time

## Stochastic Frank-Wolfe Methods for Nonconvex Optimization

Authors: S. Reddi and S. Sra and B. Poczos and A. Smola

Year: 2016

Venue: 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton-16)

abstract unavailable at this time

## Fast Incremental Method for Smooth Nonconvex Optimization

Authors: S. Reddi and S. Sra and B. Poczos and A. Smola

Year: 2016

Venue: IEEE Conference on Decision and Control (CDC-16)

abstract unavailable at this time

## Dynamical Mass Measurements of Contaminated Galaxy Clusters Using Machine Learning

Authors: M. Ntampaka and H. Trac and D. Sutherland and S. Fromenteau and B. Poczos and J. Schneider

Year: 2016

Venue: The Astrophysical Journal

We study dynamical mass measurements of galaxy clusters contaminated by interlopers and show that a modern machine learning (ML) algorithm can predict masses by better than a factor of two compared to a standard scaling relation approach. We create two mock catalogs from Multidark's publicly-available N-body MDPL1 simulation, one with perfect galaxy cluster membership information and the other where a simple cylindrical cut around the cluster center allows interlopers to contaminate the clusters. In the standard approach, we use a power law scaling relation to infer cluster mass from galaxy line of sight (LOS) velocity dispersion. Assuming perfect membership knowledge, this unrealistic case produces a wide fractional mass error distribution, with width = 0.87. Interlopers introduce additional scatter, significantly widening the error distribution further (width = 2.13). We employ the Support Distribution Machine (SDM) class of algorithms to learn from distributions of data to predict single values. Applied to distributions of galaxy observables such as LOS velocity and projected distance from the cluster center, SDM yields better than a factor-of-two improvement (width = 0.67). Remarkably, SDM applied to contaminated clusters is better able to recover masses than even the scaling relation approach applied to uncontaminated clusters. We show that the SDM method more accurately reproduces the cluster mass function, making it a valuable tool for employing cluster observations to evaluate cosmological models.

## Nonparametric Distribution Regression Applied to Sensor Modeling

Authors: A. Tallavajhula and A. Kelly and B. Poczos

Year: 2016

Venue: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-16)

abstract unavailable at this time

## Utilize Old Coordinates: Faster Doubly Stochastic Gradients for Kernel Methods

Authors: C. Li and B. Poczos

Year: 2016

Venue: Uncertainty in Artificial Intelligence (UAI-16)

abstract unavailable at this time

## Estimating Cosmological Parameters from the Dark Matter Distribution

Authors: M. Ravanbakhsh and J. Oliva and S. Fromenteau and L. Price and S. Ho and J. Schneider and B. Poczos

Year: 2016

Venue: International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Stochastic Variance Reduction for Nonconvex Optimization

Authors: S. Reddi and A. Hefny and S. Sra and B. Poczos and A. Smola

Year: 2016

Venue: International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Boolean Matrix Factorization and Noisy Completion via Message Passing

Authors: M. Ravanbakhsh and B. Poczos and R. Greiner

Year: 2016

Venue: International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Nonparametric Risk and Stability Analysis for Multi-Task Learning Problems

Authors: X. Wang and J. Oliva and J. Schneider and B. Poczos

Year: 2016

Venue: International Joint Conference on Artificial Intelligence (IJCAI)

abstract unavailable at this time

## High Dimensional Bayesian Optimization via Restricted Projection Pursuit Models

Authors: C. Li and K. Kandasamy and B. Poczos and J. Schneider

Year: 2016

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Stochastic Neural Networks with Monotonic Activation Functions

Authors: M. Ravanbakhsh and D. Schuurmans and R. Greiner and B. Poczos and J. Schneider

Year: 2016

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Bayesian Nonparametric Kernel-Learning

Authors: J. Oliva and A. Dubey and A. Wilson and B. Poczos and J. Schneider and E. Xing

Year: 2016

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Linear-time Learning on Distributions with Approximate Kernel Embeddings

Authors: Sutherland, D. and Oliva, J. and Poczos, B. and Schneider, J.

Year: 2016

Venue: 30th AAAI Conference on Artifical Intelligence (AAAI-16)

abstract unavailable at this time

## Exploration and Evaluation of AR, MPCA and KL Anomaly Detection Techniques to Embankment Dam Piezometer Data

Authors: Jung, I. and Berges, M. and Garrett, J. and Poczos, B.

Year: 2015

abstract unavailable at this time

## High Dimensional Bayesian Optimization and Bandits via Additive Models

Authors: K. Kandasamy and J. Schneider and B. Poczos

Year: 2015

Venue: International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Bayesian Active Learning for Posterior Estimation

Authors: K. Kandasamy and J. Schneider and B. Poczos

Year: 2015

Venue: International Joint Conference on Artificial Intelligence (IJCAI)

abstract unavailable at this time

## A Machine Learning Approach for Dynamical Mass Measurements of Galaxy Clusters

Authors: M. Ntampaka and H. Trac and D. Sutherland and N. Battaglia and B. Poczos and J. Schneider

Year: 2015

Venue: The Astrophysical Journal

We present a modern machine learning approach for cluster dynamical mass measurements that is a factor of two improvement over using a conventional scaling relation. Different methods are tested against a mock cluster catalog constructed using halos with mass $\geq 10^{14} \ \mathrm{M_\odot}h^{-1}$ from Multidark's publicly-available $N$-body MDPL halo catalog. In the conventional method, we use a standard $M(\sigma_v)$ power law scaling relation to infer cluster mass, $M$, from line-of-sight (LOS) galaxy velocity dispersion, $\sigma_v$. The resulting fractional mass error distribution is broad, with width $\Delta \epsilon \approx 0.86$ (68\% scatter), and has extended high-error tails. The standard scaling relation can be simply enhanced by including higher-order moments of the LOS velocity distribution. Applying the kurtosis as a linear correction term to $\log(\sigma_v)$ reduces the width of the error distribution to $\Delta \epsilon \approx 0.74$ (15\% improvement). Machine learning can be used to take full advantage of all the information in the velocity distribution. We employ the Support Distribution Machines (SDMs) algorithm that learns from distributions of data to predict single values. SDMs trained and tested on the distribution of LOS velocities yield $\Delta \epsilon \approx 0.41$ (52\% improvement). Furthermore, the problematic tails of the mass error distribution are effectively eliminated

## On Estimating L_2^2 Divergence

Authors: A. Krishnamurthy and K. Kandasamy and B. Poczos and L. Wasserman

Year: 2015

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

We give a comprehensive theoretical characterization of a nonparametric estimator for the L_2^2 divergence between two continuous distributions. We first bound the rate of convergence of our estimator, showing that it is sqrt(n)-consistent provided the densities are sufficiently smooth. In this smooth regime, we then show that our estimator is asymptotically normal, construct asymptotic confidence intervals, and establish a Berry-Esseen style inequality characterizing the rate of convergence to normality. We also show that this estimator is minimax optimal.

## Fast Function to Function Regression

Authors: J. Oliva and W. Neiswanger and B. Poczos and E. Xing and J. Schneider

Year: 2015

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

We analyze the problem of regression when both input covariates and output responses are functions from a nonparametric function class. Function to function regression (FFR) covers a large range of interesting applications including time-series prediction problems, and also more general tasks like studying a mapping between two separate types of distributions. However, previous nonparametric estimators for FFR type problems scale badly computationally with the number of input/output pairs in a data-set. Given the complexity of a mapping between general functions it may be necessary to consider large data-sets in order to achieve a low estimation risk. To address this issue, we develop a novel scalable nonparametric estimator, the Triple-Basis Estimator (3BE), which is capable of operating over datasets with many instances. To the best of our knowledge, the 3BE is the first nonparametric FFR estimator that can scale to massive datasets. We analyze the 3BE's risk and derive an upperbound rate. Furthermore, we show an improvement of several orders of magnitude in terms of prediction speed and a reduction in error over previous estimators in various real-world data-sets.

## Two-stage Sampled Learning Theory on Distributions

Authors: Z. Szabo and A. Gretton and B. Poczos and B. Sriperumbudur

Year: 2015

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

We focus on the distribution regression problem: regressing to a real-valued response from a probability distribution. Although there exist a large number of similarity measures between distributions, very little is known about their generalization performance in specific learning tasks. Learning problems formulated on distributions have an inherent two-stage sampled difficulty: in practice only samples from sampled distributions are observable, and one has to build an estimate on similarities computed between sets of points. To the best of our knowledge, the only existing method with consistency guarantees for distribution regression requires kernel density estimation as an intermediate step (which suffers from slow convergence issues in high dimensions), and the domain of the distributions to be compact Euclidean. In this paper, we provide theoretical guarantees for a remarkably simple algorithmic alternative to solve the distribution regression problem: embed the distributions to a reproducing kernel Hilbert space, and learn a ridge regressor from the embeddings to the outputs. Our main contribution is to prove the consistency of this technique in the two-stage sampled setting under mild conditions (on separable, topological domains endowed with kernels). For a given total number of observations, we derive convergence rates as an explicit function of the problem difficulty. As a special case, we answer a 15-year-old open question: we establish the consistency of the classical set kernel [Haussler, 1999; Gartner et. al, 2002] in regression, and cover more recent kernels on distributions, including those due to [Christmann and Steinwart, 2010].

## On the High-dimensional Power of Linear-time Kernel Two-Sample Testing under Mean-difference Alternatives

Authors: A. Ramdas and S. Reddi and A. Singh and B. Poczos and L. Wasserman

Year: 2015

Venue: International Conference on Artificial Intelligence and Statistics (AISTATS)

Nonparametric two sample testing deals with the question of consistently deciding if two distributions are different, given samples from both, without making any parametric assumptions about the form of the distributions. The current literature is split into two kinds of tests - those which are consistent without any assumptions about how the distributions may differ (general alternatives), and those which are designed to specifically test easier alternatives, like a difference in means (mean-shift alternatives). The main contribution of this paper is to explicitly characterize the power of a popular nonparametric two sample test, designed for general alternatives, under a mean-shift alternative in the high-dimensional setting. Specifically, we explicitly derive the power of the linear-time Maximum Mean Discrepancy statistic using the Gaussian kernel, where the dimension and sample size can both tend to infinity at any rate, and the two distributions differ in their means. As a corollary, we find that if the signal-to-noise ratio is held constant, then the test's power goes to one if the number of samples increases faster than the dimension increases. This is the first explicit power derivation for a general nonparametric test in the high-dimensional setting, and also the first analysis of how tests designed for general alternatives perform when faced with easier ones.

## Exponential Concentration of a Density Functional Estimator

Authors: S. Singh and B. Poczos

Year: 2014

Venue: Proceedings of the Neural Information Processing Systems (NIPS-14)

We analyse a plug-in estimator for a large class of integral functionals of one or more continuous probability densities. This class includes important families of entropy, divergence, mutual information, and their conditional versions. For densities on the d-dimensional unit cube [0,1]^d that lie in a beta-Holder smoothness class, we prove our estimator converges at the rate O(n^(1/(beta+d))). Furthermore, we prove that the estimator obeys an exponential concentration inequality about its mean, whereas most previous related results have bounded only expected error of estimators. Finally, we demonstrate our bounds to the case of conditional Renyi mutual information.

## On the Decreasing Power of Kernel and Distance based Nonparametric Hypothesis Tests in High Dimensions

Authors: S. Reddi and A. Ramdas and B. Poczos and A. Singh and L. Wasserman

Year: 2015

Venue: AAAI Conference on Artifical Intelligence (AAAI-15)

This paper is about two related decision theoretic problems, nonparametric two-sample testing and independence testing. There is a belief that two recently proposed solutions, based on kernels and distances between pairs of points, behave well in high-dimensional settings. We identify different sources of misconception that give rise to the above belief. Specifically, we differentiate the hardness of estimation of test statistics from the hardness of testing whether these statistics are zero or not, and explicitly discuss a notion of “fair” alternative hypotheses for these problems as dimension increases. We then demonstrate that the power of these tests actually drops polynomially with increasing dimension against fair alternatives. We end with some theoretical insights and shed light on the median heuristic for kernel bandwidth selection. Our work advances the current understanding of the power of modern nonparametric hypothesis tests in high dimensions.

## Nonparametric von Mises Estimators for Entropies, Divergences and Mutual Informations

Authors: K. Kandasamy and B. Poczos and L. Wasserman and J. Robins

Year: 2015

Venue: Proceedings of the Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## On Variance Reduction in Stochastic Gradient Descent and its Asynchronous Variants

Authors: S. Reddi and A. Hefny and S. Sra and B. Poczos and A. Smola

Year: 2015

Venue: Proceedings of the Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Communication Efficient Coresets for Empirical Loss Minimization

Authors: S. Reddi and B. Poczos and A. Smola

Year: 2015

Venue: UAI Uncertainty in Artificial Intelligence (UAI-15)

abstract unavailable at this time

## Doubly Robust Covariate Shift Correction

Authors: S. Reddi and B. Poczos and A. Smola

Year: 2015

Venue: AAAI Conference on Artifical Intelligence (AAAI-15)

Covariate shift correction allows one to perform inference even when the distribution of the covariates on the training set does not match those on the test set. This is achieved by re-weighting observations. Such a strategy removes bias, potentially at the expense of greatly increased variance. We propose a simple strategy for removing bias while retaining small variance. It uses a biased, low variance estimate as a prior and corrects the final estimate relative to the prior. We prove that this yields an efficient estimator and demonstrate good experimental performance.

## k-NN Regression on Functional Data with Incomplete Observations

Authors: S. Reddi and B. Poczos

Year: 2014

Venue: Uncertainty in Artificial Intelligence

In this paper we study a general version of regression where each covariate itself is a functional data such as distributions or functions. In real applications, however, typically we do not have direct access to such data; instead only some noisy estimates of the true covariate functions/distributions are available to us. For example, when each covariate is a distribution, then we might not be able to directly observe these distributions, but it can be assumed that i.i.d. sample sets from these distributions are available. In this paper we present a general framework and a k-NN based estimator for this regression problem. We prove consistency of the estimator and derive its convergence rates. We further show that the proposed estimator can adapt to the local intrinsic dimension in our case and provide a simple approach for choosing k. Finally, we illustrate the applicability of our framework with numerical experiments.

## An Analysis of Active Learning With Uniform Feature Noise

Authors: A. Ramdas and A. Singh and L. Wasserman and B. Poczos

Year: 2014

Venue: International Conference on AI and Statistics (AISTATS)

In active learning, the user sequentially chooses values for feature X and an oracle returns the corresponding label Y. In this paper, we consider the effect of feature noise in active learning, which could arise either because X itself is being measured, or it is corrupted in transmission to the oracle, or the oracle returns the label of a noisy version of the query point. In statistics, feature noise is known as errors in variables and has been studied extensively in non-active settings. However, the effect of feature noise in active learning has not been studied before. We consider the well-known Berkson errors-in-variables model with additive uniform noise of width \sigma. Our simple but revealing setting is that of one-dimensional binary classification setting where the goal is to learn a threshold (point where the probability of a + label crosses half). We deal with regression functions that are antisymmetric in a region of size \sigma around the threshold and also satisfy Tsybakov's margin condition around the threshold. We prove minimax lower and upper bounds which demonstrate that when \sigma is smaller than the minimiax active/passive noiseless error derived in Castro and Nowak (2007), then noise has no effect on the rates and one achieves the same noiseless rates. For larger sigma, the unflattening of the regression function on convolution with uniform noise, along with its local antisymmetry around the threshold, together yield a behaviour where noise appears to be beneficial. Our key result is that active learning can buy significant improvement over a passive strategy even in the presence of feature noise.

## Fast Distribution To Real Regression

Authors: J. Oliva and B. Poczos and J. Schneider and W. Neiswanger

Year: 2014

Venue: International Conference on AI and Statistics (AISTATS)

We study the problem of distribution to real regression, where one aims to regress a mapping f that takes in a distribution input covariate P \in I (for a non-parametric family of distributions I) and outputs a real-valued response Y = f(P) + \epsilon. This setting was recently studied in [15], where the “Kernel-Kernel” estimator was introduced and shown to have a polynomial rate of convergence. However, evaluating a new prediction with the Kernel-Kernel estimator scales as \Omega(N). This causes the difficult situation where a large amount of data may be necessary for a low estimation risk, but the computation cost of estimation becomes infeasible when the data-set is too large. To this end, we propose the Double-Basis estimator, which looks to alleviate this big data problem in two ways: First, the Double-Basis estimator is shown to have a computation complexity that is independent of the number of of instances N when evaluating new predictions after training; secondly, the Double-Basis estimator is shown to have a fast rate of convergence for a general class of mappings f in F.

## FuSSO: Functional Shrinkage and Selection Operator

Authors: J. Oliva and B. Poczos and T. Verstynen and A. Singh and J. Schneider and F. Yeh and W. Tseng

Year: 2014

Venue: International Conference on AI and Statistics (AISTATS)

We present the FuSSO, a functional analogue to the LASSO, that efficiently finds a sparse set of functional input covariates to regress a real-valued response against. The FuSSO does so in a semi-parametric fashion, making no parametric assumptions about the nature of input functional covariates and assuming a linear form to the mapping of functional covariates to the response. We provide a statistical backing for use of the FuSSO via proof of asymptotic sparsistency under various conditions. Furthermore, we observe good results on both synthetic and real-world data.

## Generalized Exponential Concentration Inequality for Renyi Divergence Estimation

Authors: S. Singh and B. Poczos

Year: 2014

Venue: International Conference on Machine Learning (ICML)

Estimating divergences in a consistent way is of great importance in many machine learning tasks. Although this is a fundamental problem in nonparametric statistics, to the best of our knowledge there has been no finite sample exponential inequality convergence bound derived for any divergence estimators. The main contribution of our work is to provide such a bound for an estimator of Renyi divergence for a smooth Holder class of densities on the d-dimensional unit cube [0, 1]^d. We also illustrate our theoretical results with a numerical experiment.

## Nonparametric Estimation of Renyi Divergence and Friends

Authors: A. Krishnamurthy and K. Kandasamy and B. Poczos and L. Wasserman

Year: 2014

Venue: International Conference on Machine Learning (ICML)

We consider nonparametric estimation of L2, Renyi-alpha and Tsallis-alpha divergences of continuous distributions. Our approach is to construct estimators for particular integral functionals of two densities and translate them into divergence estimators. For the integral functionals, our estimators are based on corrections of a preliminary plug-in estimator. We analyze the rates of convergence for our estimators and show that the parametric rate of n^{-1/2} is achievable when the densities' smoothness s are both at least d/4 where d is the dimension. We also derive minimax lower bounds for this problem which confirm that s>d/4 is necessary to achieve the n-1/2 rate of convergence. We confirm our theoretical guarantees with a number of simulations.

## Efficient Learning on Point Sets

Authors: L. Xiong and B. Poczos and J. Schneider

Year: 2013

Venue: IEEE International Conference on Data Mining (ICDM'13)

Recently several methods have been proposed to learn from data that are represented as sets of multidimensional vectors. Such algorithms usually suffer from the high demand of computational resources, making them impractical on largescale problems. We propose to solve this problem by condensing i.e. reducing the sizes of the sets while maintaining the learning performance. Three methods are examined and evaluated with a wide spectrum of set learning algorithms on several largescale image data sets. We discover that k-Means can successfully achieve the goal of condensing. In many cases, k-Means condensing can improve the algorithmsï¿½ speed, space requirements, and surprisingly, learning performances simultaneously.

## A First Look at Creating Mock Catalogs with Machine Learning Techniques

Authors: X. Xu and S. Ho and H. Trac and J. Schneider and B. Poczos and M. Ntampaka

Year: 2013

Venue: The Astrophysical Journal

We investigate machine learning (ML) techniques for predicting the number of galaxies (Ngal) that occupy a halo, given the halo's properties. These types of mappings are crucial for constructing the mock galaxy catalogs necessary for analyses of large-scale structure. The ML techniques proposed here distinguish themselves from traditional halo occupation distribution (HOD) modeling as they do not assume a prescribed relationship between halo properties and Ngal. In addition, our ML approaches are only dependent on parent halo properties (like HOD methods), which are advantageous over subhalo-based approaches as identifying subhalos correctly is difficult. We test 2 algorithms: support vector machines (SVM) and k-nearest-neighbour (kNN) regression. We take galaxies and halos from the Millennium simulation and predict Ngal by training our algorithms on the following 6 halo properties: number of particles, M200, \sigmav, vmax, half-mass radius and spin. For Millennium, our predicted Ngal values have a mean-squared-error (MSE) of ~0.16 for both SVM and kNN. Our predictions match the overall distribution of halos reasonably well and the galaxy correlation function at large scales to ~5-10%. In addition, we demonstrate a feature selection algorithm to isolate the halo parameters that are most predictive, a useful technique for understanding the mapping between halo properties and Ngal. Lastly, we investigate these ML-based approaches in making mock catalogs for different galaxy subpopulations (e.g. blue, red, high Mstar, low Mstar). Given its non-parametric nature as well as its powerful predictive and feature selection capabilities, machine learning offers an interesting alternative for creating mock catalogs.

## Active Learning and Search on Low-Rank Matrices

Authors: D. Sutherland and B. Poczos and J. Schneider

Year: 2013

Venue: Conference on Knowledge Discovery and Data Mining (KDD)

Collaborative prediction is a powerful technique, useful in domains from recommender systems to guiding the scientific discovery process. Low-rank matrix factorization is one of the most powerful tools for collaborative prediction. This work presents a general approach for active collaborative prediction with the Probabilistic Matrix Factorization model. Using variational approximations or Markov chain Monte Carlo sampling to estimate the posterior distribution over models, we can choose query points to maximize our understanding of the model, to best predict unknown elements of the data matrix, or to find as many “positive” data points as possible. We evaluate our methods on simulated data, and also show their applicability to movie ratings prediction and the discovery of drug-target interactions

## Scale Invariant Conditional Dependence Measures

Authors: S. Reddi and B. Poczos

Year: 2013

Venue: International Conference on Machine Learning (ICML)

In this paper we develop new dependence and conditional dependence measures and provide their estimators. An attractive property of these measures and estimators is that they are invariant to any monotone increasing transformations of the random variables, which is important in many applications including feature selection. Under certain conditions we show the consistency of these estimators, derive upper bounds on their convergence rates, and show that the estimators do not suffer from the curse of dimensionality. However, when the conditions are less restrictive, we derive a lower bound which proves that in the worst case the convergence can be arbitrarily slow similarly to some other estimators. Numerical illustrations demonstrate the applicability of our method.

## Distribution to Distribution Regression

Authors: J. Oliva and B. Poczos and J. Schneider

Year: 2013

Venue: International Conference on Machine Learning (ICML)

We analyze 'Distribution to Distribution regression' where one is regressing a mapping where both the covariate (inputs) and response (outputs) are distributions. No parameters on the input or output distributions are assumed, nor are any strong assumptions made on the measure from which input distributions are drawn from. We develop an estimator and derive an upper bound for the L2 risk; also, we show that when the effective dimension is small enough (as measured by the doubling dimension), then the risk converges to zero with a polynomial rate.

## Distribution-free Distribution Regression

Authors: B. Poczos and A. Rinaldo and A. Singh and L. Wasserman

Year: 2013

Venue: International Conference on AI and Statistics (AISTATS)

'Distribution regression' refers to the situation where a response Y depends on a covariate P where P is a probability distribution. The model is Y=f(P) + mu where f is an unknown regression function and mu is a random error. Typically, we do not observe P directly, but rather, we observe a sample from P. In this paper we develop theory and methods for distribution-free versions of distribution regression. This means that we do not make distributional assumptions about the error term mu and covariate P. We prove that when the effective dimension is small enough (as measured by the doubling dimension), then the excess prediction risk converges to zero with a polynomial rate.

## Nonparametric Divergence Estimation and its Applications to Machine Learning

Authors: B. Poczos and L. Xiong and J. Schneider

Year: 2012

Low-dimensional embedding, manifold learning, clustering, classification, and anomaly detection are among the most important problems in machine learning. Here we consider the setting where each instance of the inputs corresponds to a continuous probability distribution. These distributions are unknown to us, but we are given some i.i.d. samples from each of them. While most of the existing machine learning methods operate on points, i.e. finite-dimensional feature vectors, in our setting we study algorithms that operate on groups, i.e. sets of feature vectors. For this purpose, we propose new nonparametric, consistent estimators for a large family of divergences and describe how to apply them for machine learning problems. As important special cases, the estimators can be used to estimate Renyi, Tsallis, Kullback-Leibler, Hellinger, Bhattacharyya distance, L2 divergences, and mutual information. We present empirical results on synthetic data, real word images, and astronomical data sets.

## Copula-based Kernel Dependency Measures

Authors: B. Poczos and Z. Ghahramani and J. Schneider

Year: 2012

Venue: International Conference on Machine Learning (ICML)

The paper presents a new copula based method for measuring dependence between random variables. Our approach extends the Maximum Mean Discrepancy to the copula of the joint distribution. We prove that this approach has several advantageous properties. Similarly to Shannon mutual information, the proposed dependence measure is invariant to any strictly increasing transformation of the marginal variables. This is important in many applications, for example in feature selection. The estimator is consistent, robust to outliers, and uses rank statistics only. We derive upper bounds on the convergence rate and propose independence tests too. We illustrate the theoretical contributions through a series of experiments in feature selection and low-dimensional embedding of distributions.

## Nonparametric Kernel Estimators for Image Classification

Authors: B. Poczos and L. Xiong and D. Sutherland and J. Schneider

Year: 2012

Venue: 25th IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

We introduce a new discriminative learning method for image classification. We assume that the images are represented by unordered, multi-dimensional, finite sets of feature vectors, and that these sets might have different cardinality. By means of consistent nonparametric divergence estimators we define new kernels over these sets, and then apply them in kernel classifiers. Our numerical results demonstrate that in many cases this approach can outperform state-of-the-art competitors on both simulated and challenging real-world datasets.

## Conditional Distance Variance and Correlation

Authors: B. Poczos and J. Schneider

Year: 2012

Recently a new dependence measure, the distance correlation, has been proposed to measure the dependence between continuous random variables. A nice property of this measure is that it can be consistently estimated with the empirical average of the products of certain distances between the sample points. Here we generalize this quantity to measure the conditional dependence between random variables, and show that this can also be estimated with a statistic using a weighted empirical average of the products of distances between the sample points. We demonstrate the applicability of the estimators with numerical experiments on real and simulated data sets.

## Anomaly Detection for Astronomical Data

Authors: L. Xiong and B. Poczos and A. Connolly and J. Schneider

Year: 2011

Modern astronomical observatories can produce massive amount of data that are beyond the capability of the researchers to even take a glance. These scientific observations present both great opportunities and challenges for astronomers and machine learning researchers. In this project we address the problem of detecting anomalies/novelties in these large-scale astronomical data sets. Two types of anomalies, the point anomalies and the group anomalies, are considered. The point anomalies include individual anomalous objects, such as single stars or galaxies that present unique characteristics. The group anomalies include anomalous groups of objects, such as unusual clusters of the galaxies that are close together. They both have great values for astronomical studies, and our goal is to detect them automatically in un-supervised ways. For point anomalies, we adopt the subspace-based detection strategy and proposed a robust low-rank matrix decomposition algorithm for more reliable results. For group anomalies, we use hierarchical probabilistic models to capture the generative mechanism of the data, and then score the data groups using various probability measures. Experimental evaluation on both synthetic and real world data sets shows the effectiveness of the proposed methods. On a real astronomical data sets, we obtained several interesting anecdotal results. Initial inspections by the astronomers confirm the usefulness of these machine learning methods in astronomical research.

## Support Distribution Machines

Authors: B. Poczos and L. Xiong and D. Sutherland and J. Schneider

Year: 2012

Most machine learning algorithms, such as classification or regression, treat the individual data point as the object of interest. Here we consider extending machine learning algorithms to operate on groups of data points. We suggest treating a group of data points as a set of i.i.d. samples from an underlying feature distribution for the group. Our approach is to generalize kernel machines from vectorial inputs to i.i.d. sample sets of vectors. For this purpose, we use a nonparametric estimator that can consistently estimate the inner product and certain kernel functions of two distributions. The projection of the estimated Gram matrix to the cone of semi-definite matrices enables us to employ the kernel trick, and hence use kernel machines for classification, regression, anomaly detection, and low-dimensional embedding in the space of distributions. We present several numerical experiments both on real and simulated datasets to demonstrate the advantages of our new approach.

## Nonparametric Estimation of Conditional Information and Divergences

Authors: B. Poczos and J. Schneider

Year: 2012

Venue: International Conference on AI and Statistics (AISTATS)

In this paper we propose new nonparametric estimators for a family of conditional mutual information and divergences. Our estimators are easy to compute; they only use simple k nearest neighbor based statistics. We prove that the proposed conditional information and divergence estimators are consistent under certain conditions, and demonstrate their consistency and applicability by numerical experiments on simulated and real data.

## Collaborative Filtering via Online Group-structured Dictionary Learning

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2012

Venue: International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA)

Structured sparse coding and the related structured dictionary learning problems are novel research areas in machine learning. In this paper we present a new application of structured dictionary learning for collaborative filtering based recommender systems. Our extensive numerical experiments demonstrate that the presented method outperforms its state-of-the-art competitors and has several advantages over approaches that do not put structured constraints on the dictionary elements.

## Nonparametric Divergence Estimators for Independent Subspace Analysis

Authors: B. Poczos and Z. Szabo and J. Schneider

Year: 2011

Venue: The 19th European Signal Processing Conference

In this paper we propose new nonparametric Renyi, Tsallis, and L2 divergence estimators and demonstrate their applicability to mutual information estimation and independent subspace analysis. Given two independent and identically distributed samples, a naive divergence estimation approach would simply estimate the underlying densities, and plug these densities into the corresponding integral formulae. In contrast, our estimators avoid the need to consistently estimate these densities, and still they can lead to consistent estimations. Numerical experiments illustrate the efficiency of the algorithms.

## Nonparametric Independent Process Analysis

Authors: Z. Szabo and B. Poczos

Year: 2011

Venue: The 19th European Signal Processing Conference

Linear dynamical systems are widely used tools to model stochastic time processes, but they have severe limitations; they assume linear dynamics with Gaussian driving noise. Independent component analysis (ICA) aims to weaken these limitations by allowing independent, non-Gaussian sources in the model. Independent subspace analysis (ISA), an important generalization of ICA, has proven to be successful in many source separation applications. Still, the general ISA problem of separating sources with nonparametric dynamics has been hardly touched in the literature yet. The goal of this paper is to extend ISA to the case of (i) nonparametric, asymptotically stationary source dynamics and (ii) unknown source component dimensions. We make use of functional autoregressive (fAR) processes to model the temporal evolution of the hidden sources. An extension of the well-known ISA separation principle is derived for the solution of the introduced fAR independent process analysis (fAR-IPA) task. By applying fAR identification we reduce the problem to ISA. The Nadaraya-Watson kernel regression technique is adapted to obtain strongly consistent fAR estimation. We illustrate the efficiency of the fAR-IPA approach by numerical examples and demonstrate that in this framework our method is superior to standard linear dynamical system based estimators.

## ICA and ISA Using Schweizer-Wolff Measure of Dependence

Authors: S. Kirshner and B. Poczos

Year: 2008

Venue: International Conference on Machine Learning (ICML)

We propose a new algorithm for independent component and independent subspace analysis problems. This algorithm uses a contrast based on the Schweizer-Wolff measure of pairwise dependence (Schweizer & Wolff, 1981), a non-parametric measure computed on pairwise ranks of the variables. Our algorithm frequently outperforms state of the art ICA methods in the normal setting, is significantly more robust to outliers in the mixed signals, and performs well even in the presence of noise. Our method can also be used to solve independent subspace analysis (ISA) problems by grouping signals recovered by ICA methods. We provide an extensive empirical evaluation using simulated, sound, and image data.

## Budgeted Distribution Learning of Belief Net Parameters

Authors: L. Li and B. Poczos and Cs. Szepesvari and R. Greiner

Year: 2010

Venue: Proceedings of the 27th International Conference on Machine Learning

Most learning algorithms assume that a training dataset is given initially. We address the common situation where data is not available initially, but can be obtained, at a cost. We focus on learning Bayesian belief networks (BNs) over discrete variables. As such BNs are models of probabilistic distributions, we consider the generative challenge of learning the parameters for a fixed structure, that best match the true distribution. We focus on the budgeted learning setting, where there is a known fixed cost c_i for acquiring the value of the i-th feature for any specified instance, and a known total budget to spend acquiring all information. After formally defining this problem from a Bayesian perspective, we first consider non-sequential algorithms that must decide, before seeing any results, which features of which instances to probe. We show this is NP-hard, even if all variables are independent, then prove that the greedy allocation algorithm IGA is optimal here when the costs are uniform, but can otherwise be sub-optimal. We then show that general (sequential) policies perform better than non-sequential, and explore the challenges of learning the parameters for general belief networks in this sequential setting, describing conditions for when the obvious round-robin algorithm will, versus will not, work optimally. We also explore the effectiveness of this and various other heuristic algorithms.

## Budgeted Distribution Learning of Belief Net Parameters

Authors: L. Li and B. Poczos and Cs. Szepesvari and R. Greiner

Year: 2010

Most learning algorithms assume that a training dataset is given initially. We address the common situation where data is not available initially, but can be obtained, at a cost. We focus on learning Bayesian belief networks (BNs) over discrete variables. As such BNs are models of probabilistic distributions, we consider the generative challenge of learning the parameters for a fixed structure, that best match the true distribution. We focus on the budgeted learning setting, where there is a known fixed cost c_i for acquiring the value of the i-th feature for any specified instance, and a known total budget to spend acquiring all information. After formally defining this problem from a Bayesian perspective, we first consider non-sequential algorithms that must decide, before seeing any results, which features of which instances to probe. We show this is NP-hard, even if all variables are independent, then prove that the greedy allocation algorithm IGA is optimal here when the costs are uniform, but can otherwise be sub-optimal. We then show that general (sequential) policies perform better than non-sequential, and explore the challenges of learning the parameters for general belief networks in this sequential setting, describing conditions for when the obvious round-robin algorithm will, versus will not, work optimally. We also explore the effectiveness of this and various other heuristic algorithms.

## Cost Component Analysis

Authors: A. Lorincz and B. Poczos

Year: 2003

Venue: International Journal of Neural Systems

In optimizations the dimension of the problem may severely, sometimes exponentially increase optimization time. Parametric function approximatiors (FAPPs) have been suggested to overcome this problem. Here, a novel FAPP, cost component analysis (CCA) is described. In CCA, the search space is resampled according to the Boltzmann distribution generated by the energy landscape. That is, CCA converts the optimization problem to density estimation. Structure of the induced density is searched by independent component analysis (ICA). The advantage of CCA is that each independent ICA component can be optimized separately. In turn, (i) CCA intends to partition the original problem into subproblems and (ii) separating (partitioning) the original optimization problem into subproblems may serve interpretation. Most importantly, (iii) CCA may give rise to high gains in optimization time. Numerical simulations illustrate the working of the algorithm.

## Neural Kalman-filter

Authors: G. Szirtes and B. Poczos and A. Lorincz

Year: 2005

Venue: Neurocomputing, Spec. Issue

Anticipating future events is a crucial function of the central nervous system and can be modelled by Kalman filter-like mechanisms, which are optimal for predicting linear dynamical systems. Connectionist representation of such mechanisms with Hebbian learning rules has not yet been derived. We show that the recursive prediction error method offers a solution that can be mapped onto the entorhinal-hippocampal loop in a biologically plausible way. Model predictions are provided.

## Neural Kalman-filter

Authors: G. Szirtes and B. Poczos and A. Lorincz

Year: 2005

Anticipating future events is a crucial function of the central nervous system and can be modelled by Kalman filter-like mechanisms, which are optimal for predicting linear dynamical systems. Connectionist representation of such mechanisms with Hebbian learning rules has not yet been derived. We show that the recursive prediction error method offers a solution that can be mapped onto the entorhinal-hippocampal loop in a biologically plausible way. Model predictions are provided.

## Separation Theorem for K-independent Subspace Analysis with Sufficient Conditions

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2006

Here, a Separation Theorem about K-Independent Subspace Analysis (K real or complex), a generalization of K-Independent Component Analysis (KICA) is proven. According to the theorem, KISA estimation can be executed in two steps under certain conditions. In the first step, 1-dimensional KICA estimation is executed. In the second step, optimal permutation of the KICA elements is searched for. We present sufficient conditions for the KISA Separation Theorem. Namely, we shall show that (i) spherically symmetric sources (both for real and complex cases), as well as (ii) real 2-dimensional sources invariant to 90 degree rotation, among others, satisfy the conditions of the theorem.

## Finding Structure by Entropy Minimization in Coupled Reconstruction Networks

Authors: B. Szatmary and B. Poczos and A. Lorincz

Year: 2004

There is psychological and physiological evidence for components-based representations in the brain. We present a special architecture of coupled parallel working reconstruction subnetworks that can learn components of input and extract the structure of these components. Each subnetwork directly minimizes the reconstruction error and indirectly minimizes the entropy of the internal representation via a novel tuning method, which actively reduces the search space by dynamically changing learning rates and increases the escape probability from local minima. Coupled networks can reveal the structure of the input when competitive spiking and indirect minimization of the entropy of spike rate are applied together.

## Finding Structure by Entropy Minimization in Coupled Reconstruction Networks

Authors: B. Szatmary and B. Poczos and A. Lorincz

Year: 2004

Venue: Journal of Physiology

There is psychological and physiological evidence for components-based representations in the brain. We present a special architecture of coupled parallel working reconstruction subnetworks that can learn components of input and extract the structure of these components. Each subnetwork directly minimizes the reconstruction error and indirectly minimizes the entropy of the internal representation via a novel tuning method, which actively reduces the search space by dynamically changing learning rates and increases the escape probability from local minima. Coupled networks can reveal the structure of the input when competitive spiking and indirect minimization of the entropy of spike rate are applied together.

## Noncombinatorial Estimation of Independent Auto-regressive Sources

Authors: B. Poczos and A. Lorincz

Year: 2006

Venue: Neurocomputing

Identification of mixed independent subspaces is thought to suffer from combinatorial explosion of two kinds: the minimization of mutual information between the estimated subspaces and the search for the optimal number and dimensions of the subspaces. Here we show that independent auto-regressive process analysis, under certain conditions, can avoid this problem using a two-phase estimation process. We illustrate the solution by computer demonstration.

## Ockham's Razor at Work: Modeling of the Homunculus

Authors: A. Lorincz and B. Poczos and G. Szirtes and B. Takacs

Year: 2002

Venue: Brain and Mind

There is a broad consensus about the fundamental role of the hippocampal system (hippocampus and its adjacent areas) in the encoding and retrieval of episodic memories. This paper presents a functional model of this system. Although memory is not a single-unit cognitive function, we took the view that the whole system of the smooth, interrelated memory processes may have a common basis. That is why we follow the Ockham's razor principle and minimize the size or complexity of our model assumption set. The fundamental assumption is the requirement of solving the so called homunculus fallacy, which addresses the issue of interpreting the input. Generative autoassociators seem to offer a resolution of the paradox. Learning to represent and to recall information, in these generative networks, imply maximization of information transfer, sparse representation and novelty recognition. A connectionist architecture, which integrates these aspects as model constraints, is derived. Numerical studies demonstrate the novelty recognition and noise filtering properties of the architecture. Finally, we conclude that the derived connectionist architecture can be related to the neurobiological substrate.

## Fast Multidimensional Independent Component Analysis

Authors: B. Poczos and A. Lorincz

Year: 2004

A novel generalization of the well-known FastICA aglorithm, FastMICA is introduced here. The goal of FastMICA is to find hidden independent subspaces using a particular objective function family.

## Estimation of Renyi Entropy and Mutual Information Based on Generalized Nearest-Neighbor Graphs

Authors: D. Pal and B. Poczos and Cs. Szepesvari

Year: 2010

We present simple and computationally efficient nonparametric estimators of Renyi entropy and mutual information based on an i.i.d. sample drawn from an unknown, absolutely continuous distribution over R^d. The estimators are calculated as the sum of p-th powers of the Euclidean lengths of the edges of the generalized nearest-neighbor graph of the sample and the empirical copula of the sample respectively. For the first time, we prove the almost sure consistency of these estimators and upper bounds on their rates of convergence, the latter of which under the assumption that the density underlying the sample is Lipschitz continuous. Experiments demonstrate their usefulness in independent subspace analysis.

## Estimation of Renyi Entropy and Mutual Information Based on Generalized Nearest-Neighbor Graphs

Authors: D. Pal and B. Poczos and Cs. Szepesvari

Year: 2010

Venue: Proceedings of the Neural Information Processing Systems (NIPS)

We present simple and computationally efficient nonparametric estimators of Renyi entropy and mutual information based on an i.i.d. sample drawn from an unknown, absolutely continuous distribution over R^d. The estimators are calculated as the sum of p-th powers of the Euclidean lengths of the edges of the generalized nearest-neighbor graph of the sample and the empirical copula of the sample respectively. For the first time, we prove the almost sure consistency of these estimators and upper bounds on their rates of convergence, the latter of which under the assumption that the density underlying the sample is Lipschitz continuous. Experiments demonstrate their usefulness in independent subspace analysis.

## REGO: Rank-based Estimation of Renyi Information Using Euclidean Graph Optimization

Authors: B. Poczos and S. Kirshner and Cs. Szepesvari

Year: 2010

Venue: Proceedings of the 13th International Conference on AI and Statistics

We propose a new method for a nonparametric estimation of Renyi and Shannon information for a multivariate distribution using a corresponding copula, a multivariate distribution over normalized ranks of the data. As the information of the distribution is the same as the negative entropy of its copula, our method estimates this information by solving a Euclidean graph optimization problem on the empirical estimate of the distribution's copula. Owing to the properties of the copula, we show that the resulting estimator of Renyi information is strongly consistent and robust. Further, we demonstrate its applicability in image registration in addition to simulated experiments.

## On the Estimation of alpha-divergences

Authors: B. Poczos and J. Schneider

Year: 2011

Venue: International Conference on AI and Statistics (AISTATS)

We propose new nonparametric, consistent Renyi-alpha and Tsallis-alpha divergence estimators for continuous distributions. Given two independent and identically distributed samples, a naive approach would be to simply estimate the underlying densities and plug the estimated densities into the corresponding formulas. Our proposed estimators, in contrast, avoid density estimation completely, estimating the divergences directly using only simple k-nearest-neighbor statistics. We are nonetheless able to prove that the estimators are consistent under certain conditions. We also describe how to apply these estimators to mutual information and demonstrate their efficiency via numerical experiments.

## Nonparametric Divergence Estimation with Applications to Machine Learning on Distributions

Authors: B. Poczos and L. Xiong and J. Schneider

Year: 2011

Venue: Uncertainty in Artificial Intelligence (UAI)

Low-dimensional embedding, manifold learning, clustering, classification, and anomaly detection are among the most important problems in machine learning. The existing methods usually consider the case when each instance has a fixed, finite-dimensional feature representation. Here we consider a different setting. We assume that each instance corresponds to a continuous probability distribution. These distributions are unknown, but we are given some i.i.d. samples from each distribution. Our goal is to estimate the distances between these distributions and use these distances to perform low-dimensional embedding, clustering/classification, or anomaly detection for the distributions. We present estimation algorithms, describe how to apply them for machine learning tasks on distributions, and show empirical results on synthetic data, real word images, and astronomical data sets.

## Online Group-structured Dictionary Learning

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2011

Venue: 24th IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

We develop a dictionary learning method which is (i) online, (ii) enables overlapping group structures with (iii) non-convex sparsity-inducing regularization and (iv) handles the partially observable case. Structured sparsity and the related group norms have recently gained widespread attention in group-sparsity regularized problems, in the case when the dictionary is assumed to be known and fixed. However, when the dictionary also needs to be learned, the problem is much more difficult. Only a few methods have been proposed to solve this problem, and they can handle two of these four desirable properties at most. To the best of our knowledge, our proposed method is the first one that possesses all of these properties. We investigate several interesting special cases of our framework, such as the online, structured, sparse non-negative matrix factorization, and demonstrate the efficiency of our algorithm with several numerical experiments.

## Auto-regressive Independent Process Analysis without Combinatorial Efforts

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2010

Venue: Pattern Analysis and Applications

We treat the problem of searching for hidden multi-dimensional independent auto-regressive processes (Auto-Regressive Independent Process Analysis, AR-IPA). Independent Subspace Analysis (ISA) can be used to solve the AR-IPA task. The so-called separation theorem simplifies the ISA task considerably: the theorem enables one to reduce the task to 1-dimensional Blind Source Separation (BSS) task followed by the grouping of the coordinates. However, the grouping of the coordinates still involves 2 types of combinatorial problems: (i) the number of the independent subspaces and their dimensions, and then (ii) the permutation of the estimated coordinates are to be determined. Here, we generalize the separation theorem. We also show a non-combinatorial procedure, which-under certain conditions-can treat these 2 combinatorial problems. Numerical simulations have been conducted. We investigate problems that fulfill sufficient conditions of the theory and also others that do not. The success of the numerical simulations indicates that further generalizations of the separation theorem may be feasible.

## Identification of Recurrent Neural Networks by Bayesian Interrogation

Techniques

Authors: B. Poczos and A. Lorincz

Year: 2009

Venue: Journal of Machine Learning Research

We introduce novel online Bayesian methods for the identification of a family of noisy recurrent neural networks (RNNs). We present Bayesian active learning techniques for stimulus selection given past experiences. In particular, we consider the unknown parameters as stochastic variables and use A-optimality and D-optimality principles to choose optimal stimuli. We derive myopic cost functions in order to maximize the information gain concerning network parameters at each time step. We also derive the A-optimal and D-optimal estimations of the additive noise that perturbs the dynamical system of the RNN. Here we investigate myopic as well as non-myopic estimations, and study the problem of simultaneous estimation of both the system parameters and the noise. Employing conjugate priors our derivations remain approximation-free and give rise to simple update rules for the online learning of the parameters. The efficiency of our method is demonstrated for a number of selected cases, including the task of controlled independent component analysis.

## Learning When to Stop Thinking and Do Something

Authors: B. Poczos and Y. Abbasi-Yadkori and Cs. Szepesvari and R. Greiner and N. Sturtevant

Year: 2009

Venue: International Conference on Machine Learning (ICML)

An anytime algorithm is capable of returning a response to the given task at essentially any time; typically the quality of the response improves as the time increases. Here, we consider the challenge of learning when we should terminate such algorithms on each of a sequence of iid tasks, to optimize the expected average reward per unit time. We provide a system for addressing this challenge, which combines the global optimizer Cross-Entropy method with local gradient ascent. This paper theoretically investigates how far the estimated gradient is from the true gradient, then empirically demonstrates that this system is effective by applying it to a toy problem, as well as on a real-world face detection task.

## Independent Subspace Analysis Using k-nearest Neighborhood Distances

Authors: B. Poczos and A. Lorincz

Year: 2005

Venue: International Conference on Artificial Neural Networks, Artificial Neural Networks: Formal Models and Their Applications, Part II

A novel algorithm called independent subspace analysis (ISA) is introduced to estimate independent subspaces. The algorithm solves the ISA problem by estimating multi-dimensional differential entropies. Two variants are examined, both of them utilize distances between the k-nearest neighbors of the sample points. Numerical simulations demonstrate the usefulness of the algorithms.

## Cross-entropy Optimization for Independent Process Analysis

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2006

Venue: The 6th International Conference on Independent Component Analysis and Blind Source Separation

We treat the problem of searching for hidden multi-dimensional independent auto-regressive processes. First, we transform the problem to Independent Subspace Analysis (ISA). Our main contribution concerns ISA. We show that under certain conditions, ISA is equivalent to a combinatorial optimization problem. For the solution of this optimization we apply the cross-entropy method. Numerical simulations indicate that the cross-entropy method can provide considerable improvements over other state-of-the-art methods.

## Independent Subspace Analysis on Innovations

Authors: B. Poczos and B. Takacs and A. Lorincz

Year: 2005

Venue: Proceedings of the 16th European Conference on Machine Learning

Independent subspace analysis (ISA) that deals with multidimensional independent sources, is a generalization of independent component analysis (ICA). However, all known ISA algorithms may become ineffective when the sources possess temporal structure. The innovation process instead of the original mixtures has been proposed to solve ICA problems with temporal dependencies. Here we show that this strategy can be applied to ISA as well. We demonstrate the idea on a mixture of 3D processes and also on a mixture of facial pictures used as twodimensional deterministic sources. ISA on innovations was able to find the original subspaces, while plain ISA was not.

## Independent Subspace Analysis Using Geodesic Spanning Trees

Authors: B. Poczos and A. Lorincz

Year: 2005

Venue: International Conference on Machine Learning (ICML)

A novel algorithm for performing Independent Subspace Analysis, the estimation of hidden independent subspaces is introduced. This task is a generalization of Independent Component Analysis. The algorithm works by estimating the multi-dimensional differential entropy. The estimation utilizes minimal geodesic spanning trees matched to the sample points. Numerical studies include (i) illustrative examples, (ii) a generalization of the cocktail-party problem to songs played by bands, and (iii) an example on mixed independent subspaces, where subspaces have dependent sources, which are pairwise independent.

## Independent Process Analysis without A Priori Dimensional Information

Authors: B. Poczos and Z. Szabo and M. Kiszlinger and A. Lorincz

Year: 2007

Venue: Independent Component Analysis and Blind Source Separation (ICA-BSS)

Recently, several algorithms have been proposed for independent subspace analysis where hidden variables are i.i.d. processes. We show that these methods can be extended to certain AR, MA, ARMA and ARIMA tasks. Central to our paper is that we introduce a cascade of algorithms, which aims to solve these tasks without previous knowledge about the number and the dimensions of the hidden processes. Our claim is supported by numerical simulations. As an illustrative application where the dimensions of the hidden variables are unknown, we search for subspaces of facial components.

## A Cross-Entropy Method that Optimizes Partially Decomposable Problems: A New Way to Interpret NMR Spectra

Authors: M. Ravanbakhsh and R. Greiner and B. Poczos

Year: 2010

Venue: Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence. Special Track on AI and Bioinformatics

Some real-world problems are partially decomposable, in that they can be decomposed into a set of coupled subproblems, that are each relatively easy to solve. However, when these sub-problem share some common variables, it is not sufficient to simply solve each sub-problem in isolation. We develop a technology for such problems, and use it to address the challenge of finding the concentrations of the chemicals that appear in a complex mixture, based on its one-dimensional 1H Nuclear Magnetic Resonance (NMR) spectrum. As each chemical involves clusters of spatially localized peaks, this requires finding the shifts for the clusters and the concentrations of the chemicals, that collectively produce the best match to the observed NMR spectrum. Here, each sub-problem requires finding the chemical concentrations and cluster shifts that can appear within a limited spectrum range; these are coupled as these limited regions can share many chemicals, and so must agree on the concentrations and cluster shifts of the common chemicals. This task motivates CEED: a novel extension to the Cross-Entropy stochastic optimization method constructed to address such partially decomposable problems. Our experimental results in the NMR task show that our CEED system is superior to other well-known optimization methods, and indeed produces the best-known results in this important, real-world application.

## Undercomplete Blind Subspace Deconvolution

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2007

Venue: Journal of Machine Learning Research

Here we introduce the blind subspace deconvolution (BSSD) problem, which is the extension of both the blind source deconvolution (BSD) and the independent subspace analysis (ISA). We treat the undercomplete BSSD (uBSSD) case. Applying temporal concatenation we reduce this problem to ISA. The associated high dimensional ISA problem can be handled by a recent technique called joint f-decorrelation (JFD). Similar decorrelation methods have been used previously for kernel independent component analysis (kernel-ICA). More precisely, the kernel canonical correlation (KCCA) technique belongs to this family and as it is shown in this paper, the kernel generalized variance (KGV) method can also be seen as a decorrelation method in the feature space. These kernel based algorithms will be adapted to the ISA task. In the numerical examples, we (i) examine how efficiently the emerging higher dimensional ISA tasks can be tackled, and (ii) explore the working and advantages of the derived kernel-ISA methods.

## Undercomplete Blind Subspace Deconvolution via Linear Prediction

Authors: Z. Szabo and B. Poczos and and A. Lorincz

Year: 2007

Venue: European Conference on Machine Learning (ECML)

We present a novel solution technique for the blind subspace deconvolution (BSSD) problem, when temporal convolution of multidimensional hidden independent components is observed and the task is to uncover the hidden components using the observation only. We carry out this task for the undercomplete case (uBSSD): we reduce the original uBSSD task via linear prediction to independent subspace analysis (ISA), which we can solve. As it has been shown recently, applying temporal concatenation can also reduce uBSSD to ISA, but the associated ISA problem can easily become 'high dimensional'. The new reduction method circumvents this dimensionality problem. We perform detailed studies on the efficiency of the proposed technique by means of numerical simulations. We have found several advantages: our method can achieve high quality estimations for smaller number of samples and it can cope with deeper temporal convolutions.

## Post Nonlinear Independent Subspace Analysis

Authors: Z. Szabo and B. Poczos and G. Szirtes and A. Lorincz

Year: 2007

Venue: International Conference on Artificial Neural Networks (ICANN)

In this paper a generalization of Post Nonlinear Independent Component Analysis (PNL-ICA) to Post Nonlinear Independent Subspace Analysis (PNL-ISA) is presented. In this framework sources to be identified can be multidimensional as well. For this generalization we prove a separability theorem: the ambiguities of this problem are essentially the same as for the linear Independent Subspace Analysis (ISA). By applying this result we derive an algorithm using the mirror structure of the mixing system. Numerical simulations are presented to illustrate the efficiency of the algorithm.

## Hidden Markov Model Finds Behavioral Patterns of Users Working with a Headmouse Driven Writing Tool

Authors: Gy. Hevizi and M. Biczo and B. Poczos and Z. Szabo and B. Takacs and A. Lorincz

Year: 2004

Venue: International Joint Conference on Neural Networks, CD-ROM Conference Proceedings

We studied user behaviors when the cursor is directed by a head in a simple control task. We used an intelligent writing tool called Dasher. Hidden Markov models (HMMs) were applied to separate behavioral patterns. We found that a similar interpretations can be given to the hidden states upon learning. It is argued that the recognition of such general application specific behavioral patterns should be of help for adaptive humancomputer interfaces.

## Non-negative Matrix Factorization Extended by Sparse Code Shrinkage

and Weight Sparsification Algorithms

Authors: B. Szatmary and B. Poczos and J. Eggert and E. Korner and A. Lorincz

Year: 2002

Venue: ECAI2002, Proceedings of the 15th European Conference on Artificial Intelligence

Properties of a novel algorithm called non-negative matrix factorization (NMF), are studied. NMF can discover substructures and can provide estimations about the presence or the absence of those, being attractive for completion of missing information. We have studied the working and learning capabilities of NMF networks. Performance was improved by adding sparse code shrinkage (SCS) algorithm to remove structureless noise. We have found that NMF performance is considerably improved by SCS noise filtering. For improving noise resistance in the learning phase, weight sparsification was studied; a sparsifying prior was applied on the NMF weight matrix. Learning capability versus noise content was measured with and without sparsifying prior. In accordance with observation made by others on independent component analysis, we have also found that weight sparsification improved learning capabilities in the presence of Gaussian noise.

## Hierarchical Probabilistic Models for Group Anomaly Detection

Authors: L. Xiong and B. Poczos and J. Schneider and A. Connolly and J. VanderPlas

Year: 2011

Venue: International Conference on AI and Statistics (AISTATS)

Statistical anomaly detection typically focuses on finding individual data point anomalies. Often the most interesting or unusual things in a data set are not odd individual points, but rather larger scale phenomena that only become apparent when groups of data points are considered. In this paper, we propose two hierarchical probabilistic models for detecting such group anomalies. We evaluate our methods on synthetic data as well as astronomical data from the Sloan Digital Sky Survey. The experimental results show that the proposed models are effective in detecting group anomalies.

## Group Anomaly Detection Using Flexible Genre Models

Authors: L. Xiong and B. Poczos and J. Schneider

Year: 2011

Venue: Neural Information Processing Systems (NIPS)

An important task in exploring and analyzing real-world data sets is to detect unusual and interesting phenomena. In this paper, we study the group anomaly detection problem. Unlike traditional anomaly detection research that focuses on data points, our goal is to discover anomalous aggregated behaviors of groups of points. For this purpose, we propose the Flexible Genre Model (FGM). FGM is designed to characterize data groups at both the point level and the group level so as to detect various types of group anomalies. We evaluate the effectiveness of FGM on both synthetic and real data sets including images and turbulence data, and show that it is superior to existing approaches in detecting group anomalies.

## Separation Theorem for Independent Subspace Analysis and its Consequences

Authors: Z. Szabo and B. Poczos and A. Lorincz

Year: 2012

Venue: Pattern Recognition

Independent component analysis (ICA)-the theory of mixed, independent, non-Gaussian sources-has a central role in signal processing, computer vision and pattern recognition. One of the most fundamental conjectures of this research field is that independent subspace analysis (ISA)-the extension of the ICA problem, where groups of sources are independent-can be solved by a simple ICA followed by grouping the ICA components. The conjecture, called ISA separation principle, (i) has been rigorously proven for some distribution types recently, (ii) forms the basis of the state-of-the art ISA solvers, (iii) enables one to estimate the unknown number and the dimensions of the sources efficiently, and (iv) can be extended to generalizations of the ISA task, such as different linear-, controlled-, post nonlinear-, complex valued-, partially observed problems, as well as to problems dealing with nonparametric source dynamics. Here, we shall review the advances on this field.

## Hidden Markov Model Finds Behavioral Patterns Of Users Working With A Headmouse Driven Writing Tool

Authors: , György Hévízi, Mihály Biczó, Barnabás Póczos, […], András Lőrincz

Year: Jul 2004

Venue: None

Abstract Unavailable At This Time.

## Competitive Spiking And Indirect Entropy Minimization Of Rate Code: Efficient Search For Hidden Components

Authors: , Botond Szatmáry, Barnabás Póczos, András Lorincz

Year: Jul 2004

Venue: Journal of Physiology-Paris

Abstract Unavailable At This Time.

## Cost Component Analysis

Authors: , András Lörincz, Barnabás Póczos

Year: Jul 2003

Venue: International Journal of Neural Systems

Abstract Unavailable At This Time.

## Kalman-Filtering Using Local Interactions

Authors: , Barnabás Póczos, András Lörincz

Year: Feb 2003

Venue: None

Abstract Unavailable At This Time.

## Ockham'S Razor At Work: Modeling Of The Homunculus''

Authors: , András Lörincz, Barnabás Póczos, Gábor Szirtes, Bálint Takács

Year: Aug 2002

Venue: Brain and Mind

Abstract Unavailable At This Time.

## Non-Negative Matrix Factorization Extended By Sparse Code Shrinkage And Weight Sparsification Non-Negative Matrix Factorization Algorithms.

Authors: , Botond Szatmáry, Barnabás Póczos, Julian Eggert, […], András Lörincz

Year: Jan 2002

Venue: None

Abstract Unavailable At This Time.

## Non-Negative Matrix Factorization Extended By Sparse Code Shrinkage And By Weight Sparsification

Authors: , B Szatmáry, B Póczos, J Eggert, […], A LHorincz

Year: Jan 2002

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Estimation Of Conditional Information And Divergences

Authors: , Barnabás Póczos, Jeff Schneider

Year:
Venue: None

Abstract Unavailable At This Time.

## Neural Adaptation Of The Kalman-Gain

Authors: , Gábor Szirtes, Barnabás Póczos, András L˝ Orincz

Year:
Venue: None

Abstract Unavailable At This Time.

## Estimation Of Renyi Entropy And Mutual Information Based On Generalized Nearest-Neighbor Graphs.

Authors: , Dávid Pál, Barnabás Póczos, Csaba Szepesvári

Year: Jan 2010

Venue: None

Abstract Unavailable At This Time.

## A Cross-Entropy Method That Optimizes Partially Decomposable Problems: A New Way To Interpret Nmr Spectra.

Authors: , Siamak (Moshen) Ravanbakhsh, Barnabás Póczos, Russell Greiner

Year: Jan 2010

Venue: None

Abstract Unavailable At This Time.

## Identification Of Recurrent Neural Networks By Bayesian Interrogation Techniques.

Authors: , Barnabás Póczos, András Lörincz

Year: Jan 2009

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Learning When To Stop Thinking And Do Something!

Authors: , Barnabás Póczos, Yasin Abbasi-Yadkori, Csaba Szepesvári, […], Nathan R. Sturtevant

Year: Jan 2009

Venue: None

Abstract Unavailable At This Time.

## D-Optimal Bayesian Interrogation For Parameter And Noise Identification Of Recurrent Neural Networks

Authors: , Barnabás Póczos, András Lörincz

Year: Jan 2008

Venue: None

Abstract Unavailable At This Time.

## Ica And Isa Using Schweizer-Wolff Measure Of Dependence

Authors: , Sergey Kirshner, Barnabás Póczos

Year: Jan 2008

Venue: None

Abstract Unavailable At This Time.

## Undercomplete Blind Subspace Deconvolution Via Linear Prediction

Authors: , Zoltán Szabó, Barnabás Póczos, András Lőrincz

Year: Sep 2007

Venue: None

Abstract Unavailable At This Time.

## Independent Process Analysis Without A Priori Dimensional Information

Authors: , Barnabas Poczos, Zoltan Szabo, Melinda Kiszlinger, Andras Lorincz

Year: Sep 2007

Venue: None

Abstract Unavailable At This Time.

## Post Nonlinear Independent Subspace Analysis

Authors: , Zoltán Szabó, Barnabás Póczos, Gábor Szirtes, András Lörincz

Year: Sep 2007

Venue: None

Abstract Unavailable At This Time.

## Undercomplete Blind Subspace Deconvolution

Authors: , Zoltán Szabó, Barnabás Póczos, András Lörincz

Year: May 2007

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Undercomplete Blind Subspace Deconvolution Via Linear Prediction.

Authors: , Zoltán Szabó, Barnabás Póczos, András Lörincz

Year: Jan 2007

Venue: None

Abstract Unavailable At This Time.

## Non-Combinatorial Estimation Of Independent Autoregressive Sources

Authors: , Barnabás Póczos, András Lörincz

Year: Oct 2006

Venue: Neurocomputing

Abstract Unavailable At This Time.

## Separation Theorem For K-Independent Subspace Analysis With Sufficient Conditions

Authors: , Zoltan Szabo, Barnabas Poczos, Andras Lorincz

Year: Sep 2006

Venue: None

Abstract Unavailable At This Time.

## Separation Theorem For Independent Subspace Analysis With Sufficient Conditions

Authors: , Zoltan Szabo, Barnabas Poczos, Andras Lorincz

Year: Aug 2006

Venue: None

Abstract Unavailable At This Time.

## Cross-Entropy Optimization For Independent Process Analysis

Authors: , Zoltán Szabó, Barnabás Póczos, András Lörincz

Year: Mar 2006

Venue: None

Abstract Unavailable At This Time.

## Independent Subspace Analysis On Innovations

Authors: , Barnabás Póczos, Bálint Takács, András Lörincz

Year: Oct 2005

Venue: None

Abstract Unavailable At This Time.

## Independent Subspace Analysis Using K-Nearest Neighborhood Distances

Authors: , Barnabás Póczos, András Lőrincz

Year: Aug 2005

Venue: None

Abstract Unavailable At This Time.

## Separation Theorem For Independent Subspace Analysis

Authors: , Zoltán Szabó, Barnabás Póczos, András Lőrincz

Year: Jun 2005

Venue: None

Abstract Unavailable At This Time.

## Neural Kalman Filter

Authors: , Gábor Szirtes, Barnabás Póczos, András Lőrincz

Year: Jun 2005

Venue: Neurocomputing

Abstract Unavailable At This Time.

## Independent Subspace Analysis Using Geodesic Spanning Trees

Authors: , Barnabás Póczos, András Lörincz

Year: Jan 2005

Venue: None

Abstract Unavailable At This Time.

## Scale Invariant Conditional Dependence Measures

Authors: , S.J. Reddi, B. Póczos

Year: Jan 2013

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Kernel Estimators For Image Classification

Authors: , Barnabas Poczos, Liang Xiong, Dougal James Sutherland, Jeff Schneider

Year: Jun 2012

Venue: None

Abstract Unavailable At This Time.

## Copula-Based Kernel Dependency Measures

Authors: , Barnabas Poczos, Zoubin Ghahramani, Jeff Schneider

Year: Jun 2012

Venue: None

Abstract Unavailable At This Time.

## Auto-Regressive Independent Process Analysis Without Combinatorial Efforts

Authors: , Zoltán Szabó, Barnabás Póczos, András Lőrincz

Year: Apr 2012

Venue: Formal Pattern Analysis & Applications

Abstract Unavailable At This Time.

## Separation Theorem For Independent Subspace Analysis And Its Consequences

Authors: , Zoltán Szabó, Barnabás Póczos, András Lörincz

Year: Apr 2012

Venue: Pattern Recognition

Abstract Unavailable At This Time.

## Collaborative Filtering Via Group-Structured Dictionary Learning

Authors: , Zoltan Szabo, Barnabas Poczos, Andras Lorincz

Year: Mar 2012

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Divergence Estimation With Applications To Machine Learning On Distributions

Authors: , Barnabás Póczos, Liang Xiong, Jeff G. Schneider

Year: Feb 2012

Venue: None

Abstract Unavailable At This Time.

## Kernels On Sample Sets Via Nonparametric Divergence Estimates

Authors: , Dougal J. Sutherland, Liang Xiong, Barnabás Póczos, Jeff Schneider

Year: Feb 2012

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Independent Process Analysis

Authors: , Zoltán Szabó, Barnabás Póczos

Year: Aug 2011

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Divergence Estimators For Independent Subspace Analysis

Authors: , Barnabás Póczos, Zoltán Szabó, Jeff Schneider

Year: Aug 2011

Venue: None

Abstract Unavailable At This Time.

## Online Dictionary Learning With Group Structure Inducing Norms

Authors: , Zoltán Szabó, Barnabás Póczos, András Lőrincz LHorincz

Year: Jul 2011

Venue: None

Abstract Unavailable At This Time.

## Online Group-Structured Dictionary Learning

Authors: , Zoltán Szabó, Barnabás Póczos, András Lörincz

Year: Jun 2011

Venue: None

Abstract Unavailable At This Time.

## On The Estimation Of Alpha-Divergences.

Authors: , Barnabás Póczos, Jeff G. Schneider

Year: Jan 2011

Venue: None

Abstract Unavailable At This Time.

## On The Estimation Of Α-Divergences

Authors: , Barnabás Póczos, Jeff Schneider

Year: Jan 2011

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Hierarchical Probabilistic Models For Group Anomaly Detection.

Authors: , Liang Xiong, Barnabás Póczos, Jeff G. Schneider, […], Jake VanderPlas

Year: Jan 2011

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Nonparametric Divergence Estimation With Applications To Machine Learning On Distributions.

Authors: , Barnabás Póczos, Liang Xiong, Jeff G. Schneider

Year: Jan 2011

Venue: None

Abstract Unavailable At This Time.

## Group Anomaly Detection Using Flexible Genre Models

Authors: , Liang Xiong, Barnabás Póczos, Jeff Schneider

Year: Jan 2011

Venue: None

Abstract Unavailable At This Time.

## Estimation Of Rényi Entropy And Mutual Information Based On Generalized Nearest-Neighbor Graphs (Extended Version)

Authors: , D Pál, B Póczos, Cs. Szepesvári

Year: Dec 2010

Venue: None

Abstract Unavailable At This Time.

## Budgeted Distribution Learning Of Belief Net Parameters

Authors: , Liuyang Li, Barnabás Póczos, Csaba Szepesvári, Russell Greiner

Year: Aug 2010

Venue: None

Abstract Unavailable At This Time.

## Rego: Rank-Based Estimation Of Renyi Information Using Euclidean Graph Optimization.

Authors: , Barnabás Póczos, Sergey Kirshner, Csaba Szepesvári

Year: Jan 2010

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Vector-Valued Distribution Regression: A Simple And Consistent Approach

Authors: , Z Szabo, A Gretton, B Poczos, B Sriperumbudur

Year: Oct 2014

Venue: None

Abstract Unavailable At This Time.

## A Machine Learning Approach For Dynamical Mass Measurements Of Galaxy Clusters

Authors: , Michelle Ntampaka, Hy Trac, Dougal J. Sutherland, […], Jeff Schneider

Year: Oct 2014

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Simple Consistent Distribution Regression On Compact Metric Domains *

Authors: , Zoltán Szabó, Arthur Gretton, Barnabás Póczos, Bharath Sriperumbudur

Year: Sep 2014

Venue: None

Abstract Unavailable At This Time.

## Simple Consistent Distribution Regression On Compact Metric Domains (Poster)

Authors: , Zoltán Szabó, Arthur Gretton, Barnabás Póczos, Bharath Sriperumbudur

Year: Aug 2014

Venue: None

Abstract Unavailable At This Time.

## Kernel Mmd, The Median Heuristic And Distance Correlation In High Dimensions

Authors: , Sashank J. Reddi, Aaditya Ramdas, Barnabás Póczos, […], Larry Wasserman

Year: Jun 2014

Venue: None

Abstract Unavailable At This Time.

## Distribution Regression - The Set Kernel Heuristic Is Consistent

Authors: , Z Szabo, A Gretton, B Póczos, B Sriperumbudur

Year: May 2014

Venue: None

Abstract Unavailable At This Time.

## Learning On Distributions

Authors: , Zoltan Szabo, Arthur Gretton, Barnabás Póczos, Bharath Sriperumbudur

Year: Apr 2014

Venue: None

Abstract Unavailable At This Time.

## Consistent Distribution Regression Via Mean Embedding

Authors: , Z Szabo, A Gretton, B Póczos, B Sriperumbudur

Year: Mar 2014

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Estimation Of Renyi Divergence And Friends

Authors: , Akshay Krishnamurthy, Kirthevasan Kandasamy, Barnabas Poczos, Larry Wasserman

Year: Feb 2014

Venue: None

Abstract Unavailable At This Time.

## Exponential Concentration Of A Density Functional Estimator

Authors: , Shashank Singh, B. Póczos

Year: Jan 2014

Venue: Advances in neural information processing systems

Abstract Unavailable At This Time.

## Generalized Exponential Concentration Inequality For R\'Enyi Divergence Estimation

Authors: , Shashank Singh, Barnabás Póczos

Year: Jan 2014

Venue: None

Abstract Unavailable At This Time.

## K-Nn Regression On Functional Data With Incomplete Observations

Authors: , S.J. Reddi, B. Póczos

Year: Jan 2014

Venue: None

Abstract Unavailable At This Time.

## Efficient Learning On Point Sets

Authors: , Liang Xiong, Barnabas Poczos, Jeff Schneider

Year: Dec 2013

Venue: None

Abstract Unavailable At This Time.

## Fusso: Functional Shrinkage And Selection Operator

Authors: , Junier B. Oliva, Barnabas Poczos, Timothy Verstynen, […], Wen-Yih Tseng

Year: Nov 2013

Venue: None

Abstract Unavailable At This Time.

## Fast Distribution To Real Regression

Authors: , Junier B. Oliva, Willie Neiswanger, Barnabas Poczos, […], Eric Xing

Year: Nov 2013

Venue: None

Abstract Unavailable At This Time.

## Active Learning And Search On Low-Rank Matrices

Authors: , Dougal J. Sutherland, Barnabás Póczos, Jeff Schneider

Year: Aug 2013

Venue: None

Abstract Unavailable At This Time.

## A First Look At Creating Mock Catalogs With Machine Learning Techniques

Authors: , Xiaoying Xu, Shirley Ho, Hy Trac, […], Michelle Ntampaka

Year: Mar 2013

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Distribution-Free Distribution Regression

Authors: , Barnabas Poczos, Alessandro Rinaldo, Aarti Singh, Larry Wasserman

Year: Feb 2013

Venue: None

Abstract Unavailable At This Time.

## Distribution To Distribution Regression

Authors: , J.B. Oliva, B. Póczos, J. Schneider

Year: Jan 2013

Venue: None

Abstract Unavailable At This Time.

## Using Machine Learning To Populate Halos With Galaxies

Authors: , Xiaoying Xu, S. Ho, M. Ntampaka, […], H. Trac

Year: Jan 2013

Venue: None

Abstract Unavailable At This Time.

## Deep Mean Maps

Authors: , Junier B. Oliva, Dougal J. Sutherland, Barnabás Póczos, Jeff Schneider

Year: Nov 2015

Venue: None

Abstract Unavailable At This Time.

## Exploration And Evaluation Of Ar, Mpca And Kl Anomaly Detection Techniques To Embankment Dam Piezometer Data

Authors: , In-Soo Jung, Mario Berges, James H. Garrett, Barnabas Poczos

Year: Oct 2015

Abstract Unavailable At This Time.

## Linear-Time Learning On Distributions With Approximate Kernel Embeddings

Authors: , Dougal J. Sutherland, Junier B. Oliva, Barnabás Póczos, Jeff Schneider

Year: Sep 2015

Venue: None

Abstract Unavailable At This Time.

## Dynamical Mass Measurements Of Contaminated Galaxy Clusters Using Machine Learning

Authors: , M. Ntampaka, H. Trac, D. J. Sutherland, […], J. Schneider

Year: Sep 2015

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Adaptivity And Computation-Statistics Tradeoffs For Kernel And Distance Based High Dimensional Two Sample Testing

Authors: , Aaditya Ramdas, Sashank J. Reddi, Barnabas Poczos, […], Larry Wasserman

Year: Aug 2015

Venue: None

Abstract Unavailable At This Time.

## Bayesian Nonparametric Kernel-Learning

Authors: , Junier Oliva, Avinava Dubey, Barnabas Poczos, […], Eric P. Xing

Year: Jun 2015

Venue: None

Abstract Unavailable At This Time.

## On Variance Reduction In Stochastic Gradient Descent And Its Asynchronous Variants

Authors: , Sashank J. Reddi, Ahmed Hefny, Suvrit Sra, […], Alex Smola

Year: Jun 2015

Venue: None

Abstract Unavailable At This Time.

## An Analysis Of Active Learning With Uniform Feature Noise

Authors: , Aaditya Ramdas, Barnabas Poczos, Aarti Singh, Larry Wasserman

Year: May 2015

Venue: None

Abstract Unavailable At This Time.

## Two-Stage Sampled Learning Theory On Distributions

Authors: , Zoltan Szabo, Arthur Gretton, Barnabas Poczos, Bharath Sriperumbudur

Year: May 2015

Venue: None

Abstract Unavailable At This Time.

## Distribution Regression - Make It Simple And Consistent

Authors: , Zoltan Szabo, Bharath K. Sriperumbudur, Barnabás Póczos, Arthur Gretton

Year: Apr 2015

Venue: None

Abstract Unavailable At This Time.

## High Dimensional Bayesian Optimisation And Bandits Via Additive Models

Authors: , Kirthevasan Kandasamy, Jeff Schneider, Barnabas Poczos

Year: Mar 2015

Venue: None

Abstract Unavailable At This Time.

## A Simple And Consistent Technique For Vector-Valued Distribution Regression

Authors: , Z Szabo, B Sriperumbudur, B Poczos, A Gretton

Year: Jan 2015

Venue: None

Abstract Unavailable At This Time.

## Consistent Vector-Valued Regression On Probability Measures

Authors: , Z Szabo, B Sriperumbudur, B Poczos, A Gretton

Year: Jan 2015

Venue: None

Abstract Unavailable At This Time.

## Consistent Vector-Valued Distribution Regression (Slides)

Authors: , Zoltán Szabó, Arthur Gretton, Barnabás Póczos, Bharath K. Sriperumbudur

Year: Jan 2015

Venue: None

Abstract Unavailable At This Time.

## Consistent Vector-Valued Distribution Regression

Authors: , Zoltán Szabó, Arthur Gretton, Barnabás Póczos, Bharath K. Sriperumbudur

Year: Jan 2015

Venue: None

Abstract Unavailable At This Time.

## Communication Efficient Coresets For Empirical Loss Minimization

Authors: , S.J. Reddi, B. Póczos, A. Smola

Year: Jan 2015

Venue: None

Abstract Unavailable At This Time.

## On The High-Dimensional Power Of Linear-Time Kernel Two-Sample Testing Under Mean-Difference Alternatives

Authors: , Aaditya Ramdas, Sashank J. Reddi, Barnabas Poczos, […], Larry Wasserman

Year: Nov 2014

Venue: None

Abstract Unavailable At This Time.

## Influence Functions For Machine Learning: Nonparametric Estimators For Entropies, Divergences And Mutual Informations

Authors: , Kirthevasan Kandasamy, Akshay Krishnamurthy, Barnabas Poczos, […], James M. Robins

Year: Nov 2014

Venue: None

Abstract Unavailable At This Time.

## On Estimating $L 2^2$ Divergence

Authors: , Akshay Krishnamurthy, Kirthevasan Kandasamy, Barnabas Poczos, Larry Wasserman

Year: Oct 2014

Venue: None

Abstract Unavailable At This Time.

## Fast Function To Function Regression

Authors: , Junier Oliva, Willie Neiswanger, Barnabas Poczos, […], Jeff Schneider

Year: Oct 2014

Venue: None

Abstract Unavailable At This Time.

## Transformation Function Based Methods For Model Shift

Authors: , Simon Shaolei Du, Jayanth Koushik, Aarti Singh, Barnabas Poczos

Year: Dec 2016

Venue: None

Abstract Unavailable At This Time.

## Fast Incremental Method For Smooth Nonconvex Optimization

Authors: , Sashank J. Reddi, Suvrit Sra, Barnabas Poczos, Alex Smola

Year: Dec 2016

Venue: None

Abstract Unavailable At This Time.

## Quantifying Differences And Similarities In Whole-Brain White Matter Architecture Using Local Connectome Fingerprints

Authors: , Fang-Cheng Yeh, Jean M. Vettel, Aarti Singh, […], Timothy D. Verstynen

Year: Nov 2016

Venue: PLoS Computational Biology

Abstract Unavailable At This Time.

## Deep Learning With Sets And Point Clouds

Authors: , Siamak Ravanbakhsh, Jeff Schneider, Barnabas Poczos

Year: Nov 2016

Venue: None

Abstract Unavailable At This Time.

## Annealing Gaussian Into Relu: A New Sampling Strategy For Leaky-Relu Rbm

Authors: , Chun-Liang Li, Siamak Ravanbakhsh, Barnabas Poczos

Year: Nov 2016

Venue: None

Abstract Unavailable At This Time.

## Query Efficient Posterior Estimation In Scientific Experiments Via Bayesian Active Learning

Authors: , Kirthevasan Kandasamy, Jeff Schneider, Barnabás Póczos

Year: Nov 2016

Venue: None

Abstract Unavailable At This Time.

## The Multi-Fidelity Multi-Armed Bandit

Authors: , Kirthevasan Kandasamy, Gautam Dasarathy, Jeff Schneider, Barnabás Póczos

Year: Oct 2016

Venue: None

Abstract Unavailable At This Time.

## Learning Theory For Distribution Regression

Authors: , Zoltan Szabo, Bharath Sriperumbudur, Barnabas Poczos, Arthur Gretton

Year: Sep 2016

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Enabling Dark Energy Science With Deep Generative Models Of Galaxy Images

Authors: , Siamak Ravanbakhsh, Francois Lanusse, Rachel Mandelbaum, […], Barnabas Poczos

Year: Sep 2016

Venue: None

Abstract Unavailable At This Time.

## Stochastic Frank-Wolfe Methods For Nonconvex Optimization

Authors: , Sashank J. Reddi, Suvrit Sra, Barnabas Poczos, Alex Smola

Year: Sep 2016

Venue: None

Abstract Unavailable At This Time.

## Aide: Fast And Communication Efficient Distributed Optimization

Authors: , Sashank J. Reddi, Jakub Konečný, Peter Richtárik, […], Alex Smola

Year: Aug 2016

Venue: None

Abstract Unavailable At This Time.

## Finite-Sample Analysis Of Fixed-K Nearest Neighbor Density Functional Estimators

Authors: , Shashank Singh, Barnabás Póczos

Year: Jun 2016

Venue: None

Abstract Unavailable At This Time.

## Minimax-Optimal Distribution Regression

Authors: , Zoltan Szabo, Bharath K. Sriperumbudur, Barnabas Poczos, Arthur Gretton

Year: Jun 2016

Venue: None

Abstract Unavailable At This Time.

## Fast Stochastic Methods For Nonsmooth Nonconvex Optimization

Authors: , Sashank J. Reddi, Suvrit Sra, Barnabas Poczos, Alex Smola

Year: May 2016

Venue: None

Abstract Unavailable At This Time.

## Efficient Nonparametric Smoothness Estimation

Authors: , Shashank Singh, Simon S. Du, Barnabás Póczos

Year: May 2016

Venue: None

Abstract Unavailable At This Time.

## Analysis Of K-Nearest Neighbor Distances With Application To Entropy Estimation

Authors: , Shashank Singh, Barnabás Póczos

Year: Mar 2016

Venue: None

Abstract Unavailable At This Time.

## Multi-Fidelity Gaussian Process Bandit Optimisation

Authors: , Kirthevasan Kandasamy, Gautam Dasarathy, Junier B. Oliva, […], Barnabas Poczos

Year: Mar 2016

Venue: None

Abstract Unavailable At This Time.

## Fast Incremental Method For Nonconvex Optimization

Authors: , Sashank J. Reddi, Suvrit Sra, Barnabas Poczos, Alex Smola

Year: Mar 2016

Venue: None

Abstract Unavailable At This Time.

## Stochastic Variance Reduction For Nonconvex Optimization

Authors: , Sashank J. Reddi, Ahmed Hefny, Suvrit Sra, […], Alex Smola

Year: Mar 2016

Venue: None

Abstract Unavailable At This Time.

## Stochastic Neural Networks With Monotonic Activation Functions

Authors: , Siamak Ravanbakhsh, Barnabas Poczos, Jeff Schneider, […], Russell Greiner

Year: Dec 2015

Venue: None

Abstract Unavailable At This Time.

## Flexible Transfer Learning under Support and Model Shift

Authors: X. Wang, J. Schneider,

Year: 2014

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Active Transfer Learning under Model Shift

Authors: X. Wang, T. Huang, J. Schneider,

Year: 2014

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Active Area Search via Bayesian Quadrature

Authors: Y. Ma, R. Garnett, J. Schneider,

Year: 2014

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Fast Distribution To Real Regression

Authors: J. Oliva, W. Neiswanger, B. Poczos, J. Schneider, E. Xing

Year: 2014

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## FuSSO: Functional Shrinkage and Selection Operator

Authors: J. Oliva, B. Poczos, T. Verstynen, A. Singh, J. Schneider

Year: 2014

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Learning from Point Sets with Observational Bias

Authors: L. Xiong, J. Schneider

Year: 2014

Venue: , Uncertainty in Artificial Intelligence (UAI)

abstract unavailable at this time

## A first look at creating mock catalogs with machine learning techniques

Authors: X. Xu, S. Ho, H. Trac, J. Schneider, B. Poczos, M. Ntampaka

Year: 2014

Venue: , The Astrophysical Journal, 772, 14

abstract unavailable at this time

## Distribution to Distribution Regression

Authors: J. Oliva, B. Poczos, J. Schneider

Year: 2013

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Spectral Learning of Hidden Markov Models from Dynamic and Static Data

Authors: T. Huang, J. Schneider

Year: 2013

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Active Search on Graphs

Authors: X. Wang, R. Garnett, J. Schneider

Year: 2013

Venue: , ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)

abstract unavailable at this time

## Active Learning and Search on Low-Rank Matrices

Authors: D. Sutherland, B. Poczos, J. Schneider

Year: 2013

Venue: , ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)

abstract unavailable at this time

## Sigma-Optimality in Active Learning on Gaussian Random Fields

Authors: Y. Ma, R. Garnett, J. Schneider

Year: 2013

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Learning Hidden Markov Models from Non-sequence Data via Tensor Decomposition

Authors: T. Huang, J. Schneider

Year: 2013

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Efficient Learining on Point Sets

Authors: L.Xiong, J. Schneider

Year: 2013

Venue: , IEEE International Conference on Data Mining (ICDM)

abstract unavailable at this time

## Copula-based Kernel Dependency Measures

Authors: B. Poczos, Z. Ghahramani, J. Schneider,

Year: 2012

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Bayesian Optimal Active Search and Surveying

Authors: R. Garnett, Y. Krishnamurthy, X. Xiong, J. Schneider, R. Mann,

Year: 2012

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Maximum Margin Output Coding

Authors: Y. Zhang, J. Schneider,

Year: 2012

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Nonparametric Estimation of Conditional Information and Divergences

Authors: B. Poczos, J. Schneider,

Year: 2012

Venue: , AISTAT

abstract unavailable at this time

## A Composite Likelihood View for Multi-Label Classification

Authors: Y. Zhang, J. Schneider,

Year: 2012

Venue: , AISTAT

abstract unavailable at this time

## Nonparametric Kernel Estimators for Image Classification

Authors: B. Poczos, L. Xiong, D. Sutherland, J. Schneider,

Year: 2012

Venue: , IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

abstract unavailable at this time

## Learning Bi-clustered Vector Autoregressive Models

Authors: T. Huang, J. Schneider,

Year: 2012

Venue: , European Conference on Machine Learning (ECML)

abstract unavailable at this time

Authors: X. Wang, R. Garnett, J. Schneider,

Year: 2012

Venue: , NIPS workshop on Bayesian Optimization and Decision Makin

abstract unavailable at this time

## Learning Auto-regressive Models from Sequence and Non-sequence Data

Authors: T. Huang, J. Schneider,

Year: 2011

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Group Anomaly Detection using Flexible Genre Models

Authors: L. Xiong, B. Poczos, J. Schneider,

Year: 2011

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Classification of Stellar Spectra with Local Linear Embedding

Authors: S. Daniel, A. Connolly, J. Schneider, J. VanderPlas, L. Xiong,

Year: 2011

Venue: , Astronomical Journal, 142:20

abstract unavailable at this time

## Nonparametric Divergence Estimation with Applications to Machine Learning on Distributions

Authors: B. Poczos, L. Xiong, J. Schneider,

Year: 2011

Venue: , Uncertainty in Artificial Intelligence (UAI)

abstract unavailable at this time

## On the Estimation of Alpha-Divergences

Authors: B. Poczos, J. Schneider,

Year: 2011

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Hierarchical Probabilistic Models for Group Anomaly Detection

Authors: L. Xiong, B. Poczos, J. Schneider, A. Connolly, J. VanderPlas

Year: 2011

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Multi-Label Output Codes using Canonical Correlation Analysis

Authors: Y. Zhang, J. Schneider,

Year: 2011

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Adapting Control Policies for Expensive Systems to Changing Environments

Authors: M. Tesch, J. Schneider, H. Choset

Year: 2011

Venue: , IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS)

abstract unavailable at this time

## Using Response Surfaces and Expected Improvement to Optimize Snake Robot Gait Parameters

Authors: M. Tesch, J. Schneider, H. Choset

Year: 2011

Venue: , IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS)

abstract unavailable at this time

## Nonparametric Divergence Estimators for Independent Subspace Analysis

Authors: B. Poczos, Z. Szabo, J. Schneider,

Year: 2011

Venue: , European Signal Processing Conference (EUSIPCO)

abstract unavailable at this time

## Learning Multiple Tasks with a Sparse Matrix-Normal Penalty

Authors: Y. Zhang, J. Schneider,

Year: 2010

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Projection Penalties: Dimension Reduction without Loss

Authors: Y. Zhang, J. Schneider,

Year: 2010

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Learning Nonlinear Dynamic Models from Non-sequenced Data

Authors: T. Huang, L. Song, J. Schneider,

Year: 2010

Venue: , Artificial Intelligence and Statistics (AISTATS)

abstract unavailable at this time

## Temporal Collaborative Filtering with Bayesian Probabilistic Tensor Factorization

Authors: L. Xiong, X. Chen, T. Huang, J. Schneider, J. Carbonell,

Year: 2010

Venue: , SIAM Data Mining (SDM)

abstract unavailable at this time

## Learning Compressible Models

Authors: Y. Zhang, J. Schneider, A. Dubrawski,

Year: 2010

Venue: , SIAM Data Mining (SDM)

abstract unavailable at this time

## Efficiently Learning the Accuracy of Labeling Sources for Selective Sampling

Authors: P. Donmez, J. Carbonell, J. Schneider,

Year: 2010

Venue: , SIAM Data Mining (SDM)

abstract unavailable at this time

## A Probabilistic Framework to Learn from Multiple Annotators with Time-Varying Accuracy

Authors: P. Donmez, J. Carbonell, J. Schneider,

Year: 2009

Venue: , International Conference on Knowledge Discovery and Data Mining (KDD)

abstract unavailable at this time

## Learning Linear Dynamical Systems without Sequence Information

Authors: T. Huang, J. Schneider,

Year: 2009

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Learning the Semantic Correlation: An Alternative Way to Gain from Unlabeled Text

Authors: Y. Zhang, J. Schneider, A. Dubrawski,

Year: 2008

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Actively Learning Level-Sets of Composite Functions

Authors: B. Bryan, J. Schneider,

Year: 2008

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Strategies for the Use of Mixture-Based Synthetic Combinatorial Libraries: Scaffold Ranking, Direct Testing In Vivo, and Enhanced Deconvolution by Computational Methods

Authors: R. Houghten, C. Pinilla, M Giulianotti, J. Appel, C. Dooley, A. Nefzi, J. Ostresh, Y. Yu, G. Maggiora, J. Medina-Franco, D. Brunner, J. Schneider,

Year: 2008

Venue: , Journal of Combinatorial Chemistry, 10 (1), 3-1

abstract unavailable at this time

## "Mapping the Cosmological Confidence Ball Surface"

Authors: B. Bryan, J. Schneider, C. Miller, R. Nichol, C. Genovese, L. Wasserman,

Year: 2007

Venue: , Astrophysical Journal, 665(1), 25-4

abstract unavailable at this time

## Efficiently Computing Minimax Expected-Size Confidence Regions

Authors: B. Bryan, B. McMahan, C. Schafer, J. Schneider,

Year: 2007

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## A Study into Detection of Bio-Events in Multiple Streams of Surveillance Data

Authors: J. Roure, A. Dubrawski, J. Schneider

Year: 2007

Venue: , Intelligence and Security Informatics: Biosurveillanc

abstract unavailable at this time

## Detecting Anomalous Records in Categorical Datasets

Authors: K. Das, J. Schneider,

Year: 2007

Venue: , International Conference on Knowledge Discovery and Data Mining (KDD)

abstract unavailable at this time

## Active Learning for Identifying Function Threshold Boundaries

Authors: B. Bryan, L. Wasserman, J. Schneider, R. Nichol, C. Miller, C. Genovese,

Year: 2005

Venue: , Neural Information Processing Systems (NIPS)

abstract unavailable at this time

## Game Theoretic Control for Robot Teams

Authors: R. Emery-Montemerlo, G. Gordon, J. Schneider, S. Thrun,

Year: 2005

Venue: , International Conference on Robotics and Automatio

abstract unavailable at this time

## Learning Opportunity Costs in Multi-Robot Market Based Planners

Authors: J. Schneider, D. Apfelbaum, D. Bagnell, R. Simmons,

Year: 2005

Venue: , International Conference on Robotics and Automatio

abstract unavailable at this time

Authors: P. Hsiung, A. Moore, D. Neill, J. Schneider,

Year: 2005

Venue: , International Conference on Intelligence Analysi

abstract unavailable at this time

## Automatic Construction of Active Appearance Models as an Image Coding Problem

Authors: S. Baker, I. Matthews, J. Schneider,

Year: 2004

Venue: , IEEE Transactions on Pattern Analysis and Machine Intelligence, v. 26, no. 1

abstract unavailable at this time

## Belief state approaches to signaling alarms in surveillance systems

Authors: K. Das, A. Moore, J. Schneider,

Year: 2004

Venue: , ACM International Conference on Knowledge Discovery and Data Mining (KDD)

abstract unavailable at this time

## Approximate Solutions for Partially Observable Stochastic Games with Common Payoffs

Authors: Rosemary Emery-Montemerlo, Geoff Gordon, Jeff Schneider, Sebastian Thrun,

Year: 2004

Venue: , Autonomous Agents and Multi-Agent Systems (AAMAS)

abstract unavailable at this time

## Semantic based Biomedical Image Indexing and Retrieval

Authors: Y. Liu, N. Lazar, W. Rothfus, F. Dellaert, A. Moore, J. Schneider, T. Kanade,

Year: 2004

Venue: , Trends and Advances in Content-Based Image and Video Retrieval, Shapiro, Kriegel, and Veltkamp, ed

abstract unavailable at this time

## Policy Search by Dynamic Programming

Authors: Drew Bagnell, Sham Kakade, Andrew Ng, Jeff Schneider,

Year: 2003

Venue: , Proceedings of Neural Information Processing Systems (NIPS)

abstract unavailable at this time

Authors: J. A. Bagnell, J. Schneider,

Year: 2003

Venue: , International Joint Conference on Artificial Intelligence (IJCAI)

abstract unavailable at this time

Authors: Jeremy Kubica, Andrew Moore, Jeff Schneider,

Year: 2003

Venue: ,The Third IEEE International Conference on Data Minin

abstract unavailable at this time

Authors: Jeremy Kubica, Andrew Moore, David Cohn, Jeff Schneider,

Year: 2003

Venue: ,Proceedings of the 2003 IJCAI Text-Mining & Link-Analysis Worksho

abstract unavailable at this time

Authors: Anna Goldenberg, Jeremy Kubica, Paul Komarek, Andrew Moore, Jeff Schneider,

Year: 2003

Venue: , KDD Workshop on Link Analysis for Detecting Complex Behavio

abstract unavailable at this time

Authors: J. Kubica, A. Moore, D. Cohn, J. Schneider,

Year: 2003

Venue: , International Conference on Machine Learning (ICML)

abstract unavailable at this time

## Active Learning in Discrete Input Spaces

Authors: J. Schneider, A. Moore,

Year: 2002

Venue: , The 34th Interface Symposium, Montreal, Quebec, Apr 17-2

abstract unavailable at this time

## Real-valued All-Dimensions search: Low-overhead rapid searching over subsets of attributes

Authors: A. Moore, J. Schneider,

Year: 2002

Venue: , Conference of Uncertainty in Artificial Intelligence (UAI)

abstract unavailable at this time

Authors: J. Kubica, A. Moore, J. Schneider, Y. Yang,

Year: 2002

Venue: , Eighteenth National Conference on Artificial Intelligence (AAAI)

abstract unavailable at this time

## Controlling the False Discovery Rate in Astrophysical Data Analysis

Authors: C. Miller, C. Genovese, R. Nichol, L. Wasserman, A. Connolly, D. Reichart, A. Hopkins, J. Schneider, A. Moore,

Year: 2001

Venue: ,Astronomical Journal,122,6,3492-35

abstract unavailable at this time

## Classification-Driven Pathological Neuroimage Retrieval Using Statistical Asymmetry Measures

Authors: Y. Liu, F. Dellaert, W.E. Rothfus, A. Moore, J. Schneider, T. Kanade

Year: 2001

Venue: Proceedings of the International Conference of Medical Image Computing and Computer Assisted Intervention (MICCAI 2001), October 14-1

abstract unavailable at this time

## Reinforcement Learning for Cooperating and Communicating Reactive Agents in Electrical Power Grids

Authors: M. Riedmiller, A. Moore, J. Schneider,

Year: 2001

Venue: in Balancing Reactivity and Social Deliberation in Multi-agent Systems, edited by M. Hannebauer, J. Wendler, E. Pagello, Springe

abstract unavailable at this time

## Autonomous Helicopter Control using Reinforcement Learning Policy Search Methods

Authors: J. Andrew Bagnell, Jeff Schneider

Year: 2001

Venue: International Conference on Robotics and Automatio

abstract unavailable at this time

## Distributed Value Functions

Authors: Jeff Schneider, Weng-Keen Wong, Andrew Moore, Martin Riedmiller

Year: 1999

Venue: , International Conference on Machine Learnin

abstract unavailable at this time

## 3-D Deformable Registration of Medical Images Using a Statistical Atlas

Authors: Mei Chen, Takeo Kanade, Dean Pomerleau, Jeff Schneider,

Year: 1999

Venue: , Second International Conference on Medical Image Computing and Computer-Assisted Interventio

abstract unavailable at this time

## Value Function Based Production Scheduling

Authors: Jeff Schneider, Justin Boyan, Andrew Moore,

Year: 1998

Venue: ,International Conference on Machine Learnin

abstract unavailable at this time

## Q2: Memory-based active learning for optimizing noisy continuous functions

Authors: Andrew Moore, Jeff Schneider, Justin Boyan, Mary Lee,

Year: 1998

Venue: ,International Conference on Machine Learnin

abstract unavailable at this time

## Efficient Locally Weighted Polynomial Regression Predictions

Authors: Andrew Moore, Jeff Schneider, Kan Deng,

Year: 1997

Venue: ,International Conference on Machine Learnin

abstract unavailable at this time

## Exploiting Model Uncertainty Estimates for Safe Dynamic Control Learning

Authors: Jeff G. Schneider,

Year: 1996

Venue: ,Neural Information Processing Systems 9 (NIPS)

abstract unavailable at this time

## Active Learning on Non-Stationary Functions

Authors: Jeff G. Schneider,

Year: 1995

Venue: , Working Notes of the AAAI Fall Symposium on Active Learnin

abstract unavailable at this time

## Memory-based Stochastic Optimization

Authors: Andrew W. Moore and Jeff G. Schneider,

Year: 1995

Venue: ,Neural Information Processing Systems 8 (NIPS)

abstract unavailable at this time

## Robot Skill Learning Through Intelligent Experimentation

Authors: Jeff G. Schneider,

Year: 1995

Venue: ,PhD Thesis, University of Rocheste

abstract unavailable at this time

## Cooperative Coaching in Robot Learning

Authors: Jeff G. Schneider and Christopher M. Brown,

Year: 1995

Venue: ,International Conference on Intelligent Robots and System

abstract unavailable at this time

## Efficient Search for Robot Skill Learning: Simulation and Reality

Authors: Jeff G. Schneider and Roger F. Gans,

Year: 1994

Venue: , International Conference on Intelligent Robots and System

abstract unavailable at this time

## Nonparametric Estimation Of Conditional Information And Divergences

Authors: , Barnabás Póczos, Jeff Schneider

Year:
Venue: None

Abstract Unavailable At This Time.

## Detection Of Disjunctive Anomalous Patterns In Multidimensional Data

Authors: , Maheshkumar Sabhnani, Artur Dubrawski, Jeff Schneider

Year:
Venue: None

Abstract Unavailable At This Time.

## On The Estimation Of Α-Divergences

Authors: , Barnabás Póczos, Jeff Schneider

Year: Jan 2011

Venue: Journal of Machine Learning Research

Abstract Unavailable At This Time.

## Machine Learning For Effective Nuclear Search And Broad Area Monitoring

Authors: , Prateek Tandon, Jeff Schneider, Artur Dubrawski, […], Adam Zagorecki

Year: Jan 2011

Venue: None

Abstract Unavailable At This Time.

## Learning Detectors Of Events In Multivariate Time Series

Authors: , Josep Roure, Artur Dubrawski, Jeff Schneider

Year: Feb 2008

Venue: AMIA … Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

Abstract Unavailable At This Time.

## Controlling The False-Discovery Rate In Astrophysical Data Analysis

Authors: , Christopher J. Miller, Christopher Genovese, Robert C. Nichol, […], and Andrew Moore

Year: Dec 2007

Venue: The Astronomical Journal

Abstract Unavailable At This Time.

## Mapping The Cosmological Confidence Ball Surface

Authors: , Brent Bryan, Jeff Schneider, Christopher J. Miller, […], Larry Wasserman

Year: Apr 2007

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Multivariate Time Series Analyses Using Primitive Univariate Algorithms

Authors: , Maheshkumar Sabhnani, Artur Dubrawski, Jeff Schneider

Year: Jan 2007

Venue: None

Abstract Unavailable At This Time.

## Scalable Detection And Optimization Of N-Ary Linkages

Authors: , Andrew Moore, Jeff Schneider, Jeremy Kubica, […], Purna Sarkar

Year: Jun 2006

Venue: None

Abstract Unavailable At This Time.

## Sdss-Rass: Next Generation Of Cluster-Finding Algorithms

Authors: , Robert C. Nichol, Chris Miller, Andy J. Connolly, […], Wolfgang Voges

Year: Feb 2006

Venue: None

Abstract Unavailable At This Time.

## Statistical Computations With Astrogrid And The Grid

Authors: , Robert C. Nichol, Garry Smith, Christopher J. Miller, […], Andrew W. Moore

Year: Dec 2005

Venue: None

Abstract Unavailable At This Time.

## Massive Science With Vo And Grids

Authors: , Robert Nichol, Garry Smith, Christopher Miller, […], Andrew Moore

Year: Dec 2005

Venue: None

Abstract Unavailable At This Time.

Authors: , Jeremy Kubica, Andrew Moore, David Cohn, Jeff Schneider

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

## Finding Underlying Connections:

Authors: , Jeremy Kubica, Andrew Moore, David Cohn, Jeff Schneider

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

Authors: , Jeremy Kubica, Andrew Moore, Jeff Schneider, Yiming Yang

Year: Jul 2002

Venue: None

Abstract Unavailable At This Time.

## Non-Parametric Inference In Astrophysics

Authors: , Larry Wasserman, Christopher J. Miller, Robert C. Nichol, […], Jeff Schneider

Year: Jan 2002

Venue: None

Abstract Unavailable At This Time.

## Vext: A Virtual Observatory Exploration Toolkit

Authors: , Jeff Schneider, Andy Connolly

Year: Dec 2001

Venue: None

Abstract Unavailable At This Time.

## Fast Algorithms And Efficient Statistics: N-Point Correlation Functions

Authors: , Andrew Moore, Andy Connolly, Chris Genovese, […], Larry Wasserman

Year: Dec 2000

Venue: None

Abstract Unavailable At This Time.

## Distributed Value Functions

Authors: , Jeff Schneider, Weng-keen Wong, Andrew Moore, Martin Riedmiller

Year: Jan 2000

Venue: None

Abstract Unavailable At This Time.

## Probabilistic Registration Of 3-D Medical Images

Authors: , Mei Chen, Takeo Kanade, Dean Pomerleau, Jeff Schneider

Year: Aug 1999

Venue: None

Abstract Unavailable At This Time.

## Cached Sufficient Statistics For Automated Mining And Discovery From Massive Data Sources

Authors: , Research Showcase, Andrew Moore, Jeff Schneider, […], Paul Komarek

Year: Jan 1999

Venue: None

Abstract Unavailable At This Time.

## Memory Based Stochastic Optimization For Validation And Tuning Of Function Approximators

Authors: , Artur Dubrawski, Jeff Schneider

Year: Dec 1996

Venue: None

Abstract Unavailable At This Time.

## Active Search On Graphs

Authors: , Xuezhi Wang, Roman Garnett, Jeff Schneider

Year: Aug 2013

Venue: None

Abstract Unavailable At This Time.

## Searching For Complex Patterns Using Disjunctive Anomaly Detection

Authors: , Maheshkumar Sabhnani, Artur Dubrawski, Jeff Schneider

Year: Apr 2013

Venue: None

Abstract Unavailable At This Time.

## A First Look At Creating Mock Catalogs With Machine Learning Techniques

Authors: , Xiaoying Xu, Shirley Ho, Hy Trac, […], Michelle Ntampaka

Year: Mar 2013

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Learning Bi-Clustered Vector Autoregressive Models

Authors: , Tzu-Kuo Huang, Jeff Schneider

Year: Sep 2012

Venue: None

Abstract Unavailable At This Time.

## Submodularity In Batch Active Learning And Survey Problems On Gaussian

Random Fields

Authors: , Yifei Ma, Roman Garnett, Jeff Schneider

Year: Sep 2012

Venue: None

Abstract Unavailable At This Time.

## Maximum Margin Output Coding

Authors: , Yi Zhang, Jeff Schneider

Year: Jun 2012

Venue: None

Abstract Unavailable At This Time.

## Bayesian Optimal Active Search And Surveying

Authors: , Roman Garnett, Yamuna Krishnamurthy, Xuehan Xiong, […], Richard Mann

Year: Jun 2012

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Kernel Estimators For Image Classification

Authors: , Barnabas Poczos, Liang Xiong, Dougal James Sutherland, Jeff Schneider

Year: Jun 2012

Venue: None

Abstract Unavailable At This Time.

## Copula-Based Kernel Dependency Measures

Authors: , Barnabas Poczos, Zoubin Ghahramani, Jeff Schneider

Year: Jun 2012

Venue: None

Abstract Unavailable At This Time.

## Protein Subcellular Location Pattern Classification In Cellular Images Using Latent Discriminative Models

Authors: , Jieyue Li, Liang Xiong, Jeff Schneider, Robert F Murphy

Year: Jun 2012

Venue: Bioinformatics

Abstract Unavailable At This Time.

## An Efficient Parameter Space Search As An Alternative To Markov Chain

Monte Carlo

Authors: , Scott F. Daniel, Andrew J. Connolly, Jeff Schneider

Year: May 2012

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Source Location Via Bayesian Aggregation Of Evidence With Mobile Sensor Data

Authors: , Prateek Tandon, Peter Huggins, Artur Dubrawski, […], Karl Nelson

Year: Apr 2012

Venue: None

Abstract Unavailable At This Time.

## Learning Auto-Regressive Models From Sequence And Non-Sequence Data

Authors: , Tzu-Kuo Huang, Jeff Schneider

Year: Mar 2012

Venue: None

Abstract Unavailable At This Time.

## A Composite Likelihood View For Multi-Label Classification

Authors: , Yi Zhang, Jeff Schneider

Year: Mar 2012

Venue: None

Abstract Unavailable At This Time.

## Kernels On Sample Sets Via Nonparametric Divergence Estimates

Authors: , Dougal J. Sutherland, Liang Xiong, Barnabás Póczos, Jeff Schneider

Year: Feb 2012

Venue: None

Abstract Unavailable At This Time.

## Classification Of Stellar Spectra With Local Linear Embedding

Authors: , Scott F. Daniel, Andrew Connolly, Jeff Schneider, […], and Liang Xiong

Year: Nov 2011

Venue: The Astronomical Journal

Abstract Unavailable At This Time.

## Classification Of Stellar Spectra With Lle

Authors: , Scott F. Daniel, Andrew J. Connolly, Jeff Schneider, […], Liang Xiong

Year: Oct 2011

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Divergence Estimators For Independent Subspace Analysis

Authors: , Barnabás Póczos, Zoltán Szabó, Jeff Schneider

Year: Aug 2011

Venue: None

Abstract Unavailable At This Time.

## Bayesian Optimal Active Search On Graphs

Authors: , Roman Garnett, Yamuna Krishnamurthy, Donghan Wang, […], Richard Mann

Year: Aug 2011

Venue: None

Abstract Unavailable At This Time.

## Group Anomaly Detection Using Flexible Genre Models

Authors: , Liang Xiong, Barnabás Póczos, Jeff Schneider

Year: Jan 2011

Venue: None

Abstract Unavailable At This Time.

## Active Search For Sparse Signals With Region Sensing

Authors: , Yifei Ma, Roman Garnett, Jeff Schneider

Year: Dec 2016

Venue: None

Abstract Unavailable At This Time.

## Deep Learning With Sets And Point Clouds

Authors: , Siamak Ravanbakhsh, Jeff Schneider, Barnabas Poczos

Year: Nov 2016

Venue: None

Abstract Unavailable At This Time.

## Query Efficient Posterior Estimation In Scientific Experiments Via Bayesian Active Learning

Authors: , Kirthevasan Kandasamy, Jeff Schneider, Barnabás Póczos

Year: Nov 2016

Venue: None

Abstract Unavailable At This Time.

## The Multi-Fidelity Multi-Armed Bandit

Authors: , Kirthevasan Kandasamy, Gautam Dasarathy, Jeff Schneider, Barnabás Póczos

Year: Oct 2016

Venue: None

Abstract Unavailable At This Time.

## Enabling Dark Energy Science With Deep Generative Models Of Galaxy Images

Authors: , Siamak Ravanbakhsh, Francois Lanusse, Rachel Mandelbaum, […], Barnabas Poczos

Year: Sep 2016

Venue: None

Abstract Unavailable At This Time.

## Sorma Prateek Poster Final

Authors: , Prateek Tandon, Peter Huggins, Artur Dubrawski, […], Karl Nelson

Year: May 2016

Venue: None

Abstract Unavailable At This Time.

## Detecting Damped Lyman-$\Alpha$ Absorbers With Gaussian Processes

Authors: , Roman Garnett, Shirley Ho, Simeon Bird, Jeff Schneider

Year: May 2016

Venue: None

Abstract Unavailable At This Time.

## Multi-Fidelity Gaussian Process Bandit Optimisation

Authors: , Kirthevasan Kandasamy, Gautam Dasarathy, Junier B. Oliva, […], Barnabas Poczos

Year: Mar 2016

Venue: None

Abstract Unavailable At This Time.

## Stochastic Neural Networks With Monotonic Activation Functions

Authors: , Siamak Ravanbakhsh, Barnabas Poczos, Jeff Schneider, […], Russell Greiner

Year: Dec 2015

Venue: None

Abstract Unavailable At This Time.

## Deep Mean Maps

Authors: , Junier B. Oliva, Dougal J. Sutherland, Barnabás Póczos, Jeff Schneider

Year: Nov 2015

Venue: None

Abstract Unavailable At This Time.

## Linear-Time Learning On Distributions With Approximate Kernel Embeddings

Authors: , Dougal J. Sutherland, Junier B. Oliva, Barnabás Póczos, Jeff Schneider

Year: Sep 2015

Venue: None

Abstract Unavailable At This Time.

## Bayesian Nonparametric Kernel-Learning

Authors: , Junier Oliva, Avinava Dubey, Barnabas Poczos, […], Eric P. Xing

Year: Jun 2015

Venue: None

Abstract Unavailable At This Time.

## On The Error Of Random Fourier Features

Authors: , Dougal J. Sutherland, Jeff Schneider

Year: Jun 2015

Venue: None

Abstract Unavailable At This Time.

## High Dimensional Bayesian Optimisation And Bandits Via Additive Models

Authors: , Kirthevasan Kandasamy, Jeff Schneider, Barnabas Poczos

Year: Mar 2015

Venue: None

Abstract Unavailable At This Time.

## Fast Function To Function Regression

Authors: , Junier Oliva, Willie Neiswanger, Barnabas Poczos, […], Jeff Schneider

Year: Oct 2014

Venue: None

Abstract Unavailable At This Time.

## A Machine Learning Approach For Dynamical Mass Measurements Of Galaxy Clusters

Authors: , Michelle Ntampaka, Hy Trac, Dougal J. Sutherland, […], Jeff Schneider

Year: Oct 2014

Venue: The Astrophysical Journal

Abstract Unavailable At This Time.

## Efficient Learning On Point Sets

Authors: , Liang Xiong, Barnabas Poczos, Jeff Schneider

Year: Dec 2013

Venue: None

Abstract Unavailable At This Time.

## Fast Distribution To Real Regression

Authors: , Junier B. Oliva, Willie Neiswanger, Barnabas Poczos, […], Eric Xing

Year: Nov 2013

Venue: None

Abstract Unavailable At This Time.

## Fusso: Functional Shrinkage And Selection Operator

Authors: , Junier B. Oliva, Barnabas Poczos, Timothy Verstynen, […], Wen-Yih Tseng

Year: Nov 2013

Venue: None

Abstract Unavailable At This Time.

## Active Learning And Search On Low-Rank Matrices

Authors: , Dougal J. Sutherland, Barnabás Póczos, Jeff Schneider

Year: Aug 2013

Venue: None

Abstract Unavailable At This Time.

## Stochastic validation for automated tuning of neural network's hyper-parameters

Authors: Dubrawski, Artur

Year: 1997

Venue: Robotics and autonomous systems

abstract unavailable at this time

## Learning the semantic correlation: An alternative way to gain from unlabeled text

Authors: Zhang, Yi and Dubrawski, Artur and Schneider, Jeff G

Year: 2009

Venue: Advances in Neural Information Processing Systems

abstract unavailable at this time

## T-Cube as an enabling technology in surveillance applications

Authors: Dubrawski, Artur and Sabhnani, Maheshkumar and Ray, Saswati and Roure, Josep and Baysek, Michael

Year: 2007

abstract unavailable at this time

## T-Cube: A data structure for fast extraction of time series from large datasets

Authors: Sabhnani, Maheshkumar and Moore, Andrew W and Dubrawski, Artur W

Year: 2007

abstract unavailable at this time

## Detection of events in multiple streams of surveillance data

Authors: Dubrawski, Artur

Year: 2011

Venue: Infectious disease informatics and biosurveillance

abstract unavailable at this time

## Rapid processing of ad-hoc queries against large sets of time series

Authors: Sabhnani, Maheshkumar and Moore, A and Dubrawski, Artur

Year: 2007

abstract unavailable at this time

## The role of data aggregation in public health and food safety surveillance

Authors: Dubrawski, Artur and Zhang, X

Year: 2010

Venue: Biosurveillance: Methods and Case Studies

abstract unavailable at this time

## Monitoring food safety by detecting patterns in consumer complaints

Authors: Dubrawski, Artur and Elenberg, Kimberly and Moore, Andrew and Sabhnani, Maheshkumar

Year: 2006

Venue: Proceedings of the National Conference on Artificial Intelligence

abstract unavailable at this time

## Projection retrieval for classification

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2012

Venue: Advances in Neural Information Processing Systems

abstract unavailable at this time

## Computational thinking

Authors: Wing, Jeannette M

Year: 2006

Venue: Communications of the ACM

abstract unavailable at this time

## Trade-offs between agility and reliability of predictions in dynamic social networks used to model risk of microbial contamination of food

Authors: Dubrawski, Artur and Sarkar, Purnamrita and Chen, Lujie

Year: 2009

Venue: Social Network Analysis and Mining, 2009. ASONAM'09. International Conference on Advances in

abstract unavailable at this time

## Efficient analytics for effective monitoring of biomedical security

Authors: Sabhnani, Maheshkumar and Neill, Daniel and Moore, Andrew and Dubrawski, Artur and Wong, W

Year: 2005

Venue: Proceedings of the International Conference on Information and Automation

abstract unavailable at this time

## Learning locomotion reflexes: A self-supervised neural system for a mobile robot

Authors: Dubrawski, Artur and Crowley, James L

Year: 1994

Venue: Robotics and Autonomous Systems

abstract unavailable at this time

## Automated detection of data entry errors in a real time surveillance system

Authors: Chen, Lujie and Dubrawski, Artur and Waidyanatha, Nuwan and Weerasinghe, Chamindu

Year: 2011

Venue: Emerg Health Threats J

abstract unavailable at this time

## Multivariate time series analyses using primitive univariate algorithms

Authors: Sabhnani, Maheshkumar and Dubrawski, Artur and Schneider, Jeff

Year: 2007

abstract unavailable at this time

## Mining intensive care vitals for leading indicators of adverse health events

Authors: Lonkar, Rajas and Dubrawski, Artur and Fiterau, Madalina and Garnett, Roman

Year: 2011

Venue: Emerg Health Threats J

abstract unavailable at this time

## Applying outbreak detection algorithms to prognostics

Authors: Dubrawski, Artur and Baysek, Michael and Mikus, Maj Shannon and McDaniel, Charles and Mowry, Bradley and Moyer, Laurel and Ostlund, John and Sondheimer, Norman and Stewart, Timothy

Year: 2007

Venue: AAAI Fall Symposium on Artificial Intelligence for Prognostics

abstract unavailable at this time

## Interactive manipulation, visualization analysis of large sets of multidimensional time series in health informatics

Authors: Dubrawski, Artur and Sabhnani, Maheshkumar and Ray, Saswati and Baysek, Michael and Chen, Lujie and Ostlund, John and Knight, Michael

Year: 2008

Venue: INFORMS

abstract unavailable at this time

## Leveraging publicly available data to discern patterns of human-trafficking activity

Authors: Dubrawski, Artur and Miller, Kyle and Barnes, Matthew and Boecking, Benedikt and Kennedy, Emily

Year: 2015

Venue: Journal of Human Trafficking

abstract unavailable at this time

## Informative projection recovery for classification, clustering and regression

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2013

Venue: Machine Learning and Applications (ICMLA), 2013 12th International Conference on

abstract unavailable at this time

## T-Cube Web Interface as a tool for detecting disease outbreaks in real-time: A pilot in India and Sri Lanka

Authors: Waidyanatha, Nuwan and Sampath, Chamindu and Dubrawski, Artur and Sabhnani, Maheshkumar and Chen, Lujie and Ganesan, M and Vincy, P

Year: 2010

Venue: Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF), 2010 IEEE RIVF International Conference on

abstract unavailable at this time

## Tuning neural networks with stochastic optimization

Authors: Dubrawski, Artur

Year: 1997

Venue: Intelligent Robots and Systems, 1997. IROS'97., Proceedings of the 1997 IEEE/RSJ International Conference on

abstract unavailable at this time

## An Entity Resolution approach to isolate instances of Human Trafficking online

Authors: Nagpal, Chirag and Miller, Kyle and Boecking, Benedikt and Dubrawski, Artur

Year: 2015

Venue: arXiv preprint arXiv:1509.06659

abstract unavailable at this time

## Automatic state discovery for unstructured audio scene classification

Authors: Ramos, Julian and Siddiqi, Sajid and Dubrawski, Artur and Gordon, Geoffrey and Sharma, Abhishek

Year: 2010

Venue: Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on

abstract unavailable at this time

## Memory-based stochastic optimization for automated tuning of neural network's high level parameters

Authors: Dubrawski, Artur

Year: 1996

Venue: 4th International Symposium on Intelligent Robotic Systems

abstract unavailable at this time

## Learning stable multivariate baseline models for outbreak detection

Authors: Siddiqi, Sajid M and Boots, Byron and Gordon, Geoffrey J and Dubrawski, Artur W

Year: 2007

abstract unavailable at this time

## Learning temporal rules to forecast events in multivariate time sequences

Authors: Guillame-Bert, Mathieu and Dubrawski, Artur

Year: 2014

Venue: 2nd Workshop on Machine Learning for Clinical Data Analysis, Healthcare and Genomics. NIPS

abstract unavailable at this time

## Dynamic network model for predicting occurrences of salmonella at food facilities

Authors: Sarkar, Purnamrita and Chen, Lujie and Dubrawski, Artur

Year: 2008

Venue: Biosurveillance and Biosecurity

abstract unavailable at this time

## Computational thinking.

Authors: Wing, Jeannette M

Year: 2011

Venue: VL/HCC

abstract unavailable at this time

## Finding Meaningful Gaps to Guide Data Acquisition for a Radiation Adjudication System.

Authors: Gisolfi, Nick and Fiterau, Madalina and Dubrawski, Artur

Year: 2015

Venue: AAAI

abstract unavailable at this time

## Simultaneous detection of radioactive sources and inference of their properties

Authors: Tandon, Prateek and Huggins, Peter and Dubrawski, Artur and Labov, Simon and Nelson, Karl

Year: 2013

Venue: IEEE Nuclear Science Symposium

abstract unavailable at this time

## Real-time biosurveillance pilot in India and Sri Lanka

Authors: Waidyanatha, Nuwan and Prashant, Suma and Ganesan, M and Dubrawski, Artur and Chen, Lujie and Baysek, Michael and Careem, Mifan and Damendra, Pradeeper and Kaluarachchi, Mahesh

Year: 2010

Venue: e-Health Networking Applications and Services (Healthcom), 2010 12th IEEE International Conference on

abstract unavailable at this time

## Real-time adaptive monitoring of vital signs for clinical alarm preemption

Authors: Fiterau, Madalina and Dubrawski, Artur and Ye, Can

Year: 2011

Venue: Emerging Health Threats Journal

abstract unavailable at this time

## A study into detection of bio-events in multiple streams of surveillance data

Authors: Roure, Josep and Dubrawski, Artur and Schneider, Jeff

Year: 2007

Venue: Intelligence and security informatics: Biosurveillance

abstract unavailable at this time

## Techniques for early warning of systematic failures of aerospace components

Authors: Dubrawski, Artur and Sondheimer, Norman

Year: 2011

Venue: Aerospace Conference, 2011 IEEE

abstract unavailable at this time

## T-Cube: Quick Response to Ad-Hoc Time Series Queries against Large Datasets

Authors: Sabhnani, Maheshkumar and Dubrawski, Artur and Moore, Andrew

Year: 2011

abstract unavailable at this time

## Do Public Events Affect Sex Trafficking Activity?

Authors: Miller, Kyle and Kennedy, Emily and Dubrawski, Artur

Year: 2016

Venue: arXiv preprint arXiv:1602.05048

abstract unavailable at this time

## On Learning from Collective Data

Authors: Xiong, Liang

Year: 2013

abstract unavailable at this time

## Using AFDL algorithm to estimate risk of positive outcomes of microbial tests at food establishments

Authors: Dubrawski, Artur and Chen, Lujie and Ostlund, John

Year: 2008

abstract unavailable at this time

## Discovering Possible Linkages between Food-borne Illness and the Food Supply Using an Interactive Analysis Tool

Authors: Dubrawski, Artur and Chen, Lujie and Sabhnani, Maheshkumar and Fedorka-Cray, Paula J and Kelley, Lynda and Gerner-Smidt, Peter and Williams, Ian and Huckabee, Mark and Dunham, Adrienne

Year: 2009

Venue: 8th Annual Conference of the International Society for Disease Surveillance

abstract unavailable at this time

## Learning specific detectors of adverse events in multivariate time series

Authors: Roure, Josep and Dubrawski, Artur and Schneider, Jeff

Year: 2007

abstract unavailable at this time

## Real-time visual analysis of microvascular blood flow for critical care

Authors: Liu, Chao and Gomez, Hernando and Narasimhan, Srinivasa and Dubrawski, Artur and Pinsky, Michael R and Zuckerbraun, Brian

Year: 2015

Venue: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

abstract unavailable at this time

## Finding gaps in data to guide development of a radiation threat adjudication system

Authors: Gisolfi, Nicholas and Fiterau, Madalina and Dubrawski, Artur and Ray, S and Labov, S and Nelson, K

Year: 2014

Venue: Symposium on Radiation Measurements and Applications

abstract unavailable at this time

## Active machine learning to increase annotation efficiency in classifying vital sign events as artifact or real alerts in continuous noninvasive monitoring

Authors: Hravnak, Marilyn and Chen, Lujie and Fiterau, Madaline and Dubrawski, Artur and Clermont, Gilles and Guillame-Bert, Mathieu and Bose, Eliezer and Pinsky, Michael R

Year: 2014

Venue: B94. PREDICTIONS AND OUTCOMES IN THE ICU

abstract unavailable at this time

## TCube Web Interface for Realtime Biosurveillance in Sri Lanka

Authors: Sabhnani, Maheshkumar and Dubrawski, Artur and Waidyanatha, Nuwan

Year: 2007

abstract unavailable at this time

## Learning with limited supervision by input and output coding

Authors: Zhang, Yi

Year: 2012

abstract unavailable at this time

## Computationally efficient scoring of activity using demographics and connectivity of entities

Authors: Dubrawski, Artur W and Ostlund, John K and Chen, Lujie and Moore, Andrew W

Year: 2010

Venue: Information Technology and Management

abstract unavailable at this time

## Computational thinking

Authors: Wing, Jeannette M

Year: 2009

Venue: J. Comput. Sci. Coll

abstract unavailable at this time

## Memory based stochastic optimization for validation and tuning of function approximators

Authors: Dubrawski, Artur and Schneider, Jeff

Year: 1997

Venue: Conference on AI and Statistics

abstract unavailable at this time

## Artificial neural network for mobile robot topological localization

Authors: Racz, Janusz and Dubrawski, Artur

Year: 1995

Venue: Robotics and autonomous systems

abstract unavailable at this time

## AHMED, Amr CMU-ML-07-116

Authors: BRYAN, Brent and COHEN, William W and DUBRAWSKI, Artur W and FALOUTSOS, Christos and FUNIAK, Stanislav and GLANCE, Natalie and GOLDENBERG, Anna and GORDON, Geoff and GUESTRIN, Carlos and GUPTA, Anupam and others

Year: 1995

abstract unavailable at this time

Authors: Herlands, William and De-Arteaga, Maria and Neill, Daniel and Dubrawski, Artur

Year: 2015

Venue: arXiv preprint arXiv:1511.04402

abstract unavailable at this time

## A Data Structure for Fast Extraction of Time Series from Large Datasets

Authors: Sabhnani, Maheshkumar and Moore, Andrew W and Dubrawski, Artur W

Year: 2007

abstract unavailable at this time

## Applied Indoor Localization: Map-based, Infrastructure-free, with Divergence Mitigation and Smoothing

Authors: MacLachlan, Robert A and Dubrawski, Artur

Year: 2010

Venue: Information Fusion (FUSION….(2012)

abstract unavailable at this time

## ZHANG, Yi CMU-ML-09-110

Authors: AHMED, Amr and ARNOLD, Andrew O and CARBONELL, Jaime G and CEN, Hao and COHEN, William W and DAS, Kaustav and DUBRAWSKI, Artur and EL-ARINI, Khalid and GUESTRIN, Carlos and HANNEKE, Steve and others

Year: 2010

abstract unavailable at this time

## Interactive Analysis of Multidimensional Data

Authors: Dubrawski, Artur

Year: 2010

abstract unavailable at this time

## ASONAM 2013 program committee

Authors: Zimmermann, Albrecht and Jorge, Alipio and Appel, Ana Paula and Carvalho, Andre and Tagarelli, Andrea and Schmidt, Andreas and Dubrawski, Artur and Selman, Bart and Van De Walle, Bartel and Chien, Been-Chian and others

Year: 2010

abstract unavailable at this time

## Canonical Autocorrelation Analysis

Authors: De-Arteaga, Maria and Dubrawski, Artur and Huggins, Peter

Year: 2015

Venue: arXiv preprint arXiv:1511.06419

abstract unavailable at this time

## Batched Lazy Decision Trees

Authors: Guillame-Bert, Mathieu and Dubrawski, Artur

Year: 2016

Venue: arXiv preprint arXiv:1603.02578

abstract unavailable at this time

## Riding an emotional roller-coaster: A multimodal study of young child’s math problem solving activities

Authors: Chen, Lujie and Li, Xin and Xia, Zhuyun and Song, Zhanmei and Morency, Louis-Philippe and Dubrawski, Artur

Year: 2016

abstract unavailable at this time

## Active Learning for Informative Projection Retrieval.

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2015

Venue: AAAI

abstract unavailable at this time

## ASONAM 2010 Program Committee

Authors: Abraham, Ajith and Agarwal, Nitin and Alani, Harith and Assent, Ira and Barbian, Guido and Biba, Marenglen and Bonchi, Francesco and Braha, Dan and Cao, Bin and Carvalho, Andre and others

Year: 2015

abstract unavailable at this time

## Conference Program Committee

Authors: Ashish, Naveen and Badia, Antonio and Burgoon, Judee and Cai, Guoray and Cao, Zhidong and Chan, Chien-Lung and Chen, You and Cheng, Xueqi and CAS, PR and Chikkagoudar, Satish and others

Year: 2015

abstract unavailable at this time

## Mining sea turtle nests: An amplitude independent feature extraction method for GPR data

Authors: Ermakov, Vladimir and Dubrawski, Artur and Hodgins, Jessica and Dohi, Tony and Savage, Anne

Year: 2012

Venue: Ground Penetrating Radar (GPR), 2012 14th International Conference on

abstract unavailable at this time

## Evolution of a Useful Autonomous System

Authors: Dubrawski, Artur and Thorne, Henry

Year: 2009

Venue: Robot Motion and Control 2009

abstract unavailable at this time

## Clustering on the Edge: Learning Structure in Graphs

Authors: Barnes, Matt and Dubrawski, Artur

Year: 2016

Venue: arXiv preprint arXiv:1605.01779

abstract unavailable at this time

## Clustering on the Edge: Learning Structure in Graphs

Authors: Barnes, Matt and Dubrawski, Artur

Year: 2016

Venue: arXiv preprint arXiv:1605.01779

abstract unavailable at this time

## Performance Bounds for Pairwise Entity Resolution

Authors: Barnes, Matt and Miller, Kyle and Dubrawski, Artur

Year: 2015

Venue: arXiv preprint arXiv:1509.03302

abstract unavailable at this time

## Computationally Efficient Scoring of Activity in Large Social Networks using Connectivity Patterns and Demographics of Entities

Authors: Dubrawski, Artur W and Ostlund, John K and Chen, Lujie and Moore, Andrew W

Year: 2015

abstract unavailable at this time

## Trade-offs in Explanatory Model Learning

Authors: Fiterau, Madalina and Dubrawski, Artur and Schneider, Jeff and Gordon, Geoff

Year: 2012

abstract unavailable at this time

## ISI 2015 Program Committee

Authors: Zhou, Lina and Kaati, Lisa and Mao, Wenji and Adams, Niall and Albanese, Massimiliano and Badia, Antonio and Bahnsen, Alejandro Correa and Brynielsson, Joel and Cai, Guoray and Calvo-Rolle, Jose Luis and others

Year: 2012

abstract unavailable at this time

Authors: Barnes, Matt and Gisolfi, Nick and Fiterau, Madalina and Dubrawski, Artur

Year: 2015

Venue: AAAI

abstract unavailable at this time

## Learning Compressible Models

Authors: Dubrawski, Yi Zhang Jeff Schneider Artur

Year: 2009

abstract unavailable at this time

## ISI 2016 Program Committee

Authors: Kahn, Latifur and Kaza, Siddharth and Patton, Mark and Adali, Sibel and Adams, Niall and Akhgar, Babak and Albanese, Massimiliano and Badia, Antonio and Bahnsen, Alejandro Correa and Brynielsson, Joel and others

Year: 2009

abstract unavailable at this time

## Scalable Detection and Optimization of N-ARY Linkages

Authors: Moore, Andrew and Schneider, Jeff and Kubica, Jeremy and Goldenberg, Anna and Dubrawski, Artur and Ostlund, John and Choi, Patrick and Komarek, Jeanie and Goode, Adam and Sarkar, Purna

Year: 2006

abstract unavailable at this time

## Bio-Sensing based Adaptive Thermal Comfort Controls

Authors: Loftness, Vivian and Choi, Joon Ho and Hartkopf, Volker and Mattern, Gerry and Dubrawski, Artur

Year: 2006

abstract unavailable at this time

## An Application of Divergence Estimation to Projection Retrieval for Semi-supervised Classification and Clustering

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2006

abstract unavailable at this time

## Learning temporal association rules on Symbolic time sequences

Authors: Lux, Mr Augustin and Ghallab, Mr Malik and Lukowicz, Mr Paul and Dubrawski, Mr Artur and CROWLEY, Mr James L

Year: 2012

abstract unavailable at this time

## Data-driven Classification of Screwdriving Operations

Authors: Aronson, Reuben M and Bhatia, Ankit and Jia, Zhenzhong and Guillame-Bert, Mathieu and Bourne, David and Dubrawski, Artur and Mason, Matthew T

Year: 2012

abstract unavailable at this time

## VIPR: An Interactive Tool for Meaningful Visualization of High-Dimensional Data

Authors: Wang, Donghan and Fiterau, Madalina and Dubrawski, Artur

Year: 2012

abstract unavailable at this time

## Explanation-Oriented Classification via Subspace Partitioning

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2012

abstract unavailable at this time

## Discovering Compact and Informative Structures through Data Partitioning

Year: 2015

abstract unavailable at this time

## Auton Lab Software Applications Training-of-Trainer Program

Authors: Prashant, Suma and Kanan, T and Waidyanatha, Nuwan and Weerakoon, Pubudini and Gayan, Chinthaka

Year: 2015

abstract unavailable at this time

## ESB\# 331, Indian Institute of Technology Madras (IITM) Chennai--600 036 Website: www. rtbi. in

Authors: Prashant, Suma and Ganesan, M

Year: 2009

abstract unavailable at this time

## Active Transfer Learning

Authors: Wang, Xuezhi

Year: 2016

abstract unavailable at this time

## CMU-ML-15-105 Discovering Compact and Informative Structures through Data Partitioning

Year: 2016

abstract unavailable at this time

## Dynamic and Personalized Risk Forecast in Step-Down Units: Implications for Monitoring Paradigms

Authors: Chen, Lujie and Ogundele, Olufunmilayo and Clermont, Gilles and Hravnak, Marilyn and Pinsky, Michael R and Dubrawski, Artur W

Year: 2016

Venue: Annals of the American Thoracic Society

abstract unavailable at this time

## Fault-Tolerant Source Detection using Bayesian Sensor Reliability Models

Authors: Tandon, P and Huggins, P and Dubrawski, A and Labov, S and Nelson, K

Year: 2016

abstract unavailable at this time

## Project Work Plan

Authors: Waidyanatha, Nuwan

Year: 2016

abstract unavailable at this time

## ASONAM 2012 Program Committee

Authors: Abell, Peter and Ackland, Robert and Adamic, Lada and Afsharchi, Mohsen and Agarwal, Nitin and Akkaya, Kemal and Alani, Harith and Arshad Ali, NIIT and Appel, Pakistan Ana Paula and Arimura, Hiroki and others

Year: 2016

abstract unavailable at this time

## Detection of Radioactive Sources Using Bayesian Aggregation of Data from Mobile Spectrometers

Authors: Tandon, Prateek and Huggins, Peter and Dubrawski, Artur and Labov, Simon and Nelson, Karl

Year: 2016

abstract unavailable at this time

## 2012 IEEE International Conference on Intelligence and Security Informatics

Authors: Cyberspace, Border

Year: 2012

abstract unavailable at this time

## Is there an information hierarchy among hemodynamic variables for early identification of occult hemorrhage?

Authors: Holder, Andre and Guillame-Bert, Mathieu and Chen, Karen and Huggins, Peter and Dubrawski, Artur and Hravnak, Marilyn and Clermont, Gilles and Pinsky, Michael

Year: 2013

Venue: Journal of Critical Care

abstract unavailable at this time

Authors: Kazuo Iwano, IBM and Katz, Japan Randy H and Khatib, Oussama and Truszkowski, Walter F and Verma, Dinesh C and Subhash Wadhva, IIT and Yousif, India Mazim S and de Souza, Jos{\'e} Neuman and Esmahi, Larbi and van der Meer, Sven and others

Year: 2013

abstract unavailable at this time

## Exploiting Non-sequence Data in Dynamic Model Learning

Authors: Huang, Tzu-Kuo

Year: 2013

abstract unavailable at this time

## Rule-Based Anomaly Pattern Detection for Detecting Disease Outbreaks

Authors: Moore, Andrew and Wong, Weng-Keen and Wagner, Mike and Cooper, Greg

Year: 2003

Venue: Proceedings of the 35th Symposium on the Interface of Computing and Statistics

abstract unavailable at this time

## An Application of Divergence Estimation to Projection Retrieval for Semi-supervised Classification and Clustering

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2003

abstract unavailable at this time

## Learning temporal association rules on Symbolic time sequences

Authors: Lux, Mr Augustin and Ghallab, Mr Malik and Lukowicz, Mr Paul and Dubrawski, Mr Artur and CROWLEY, Mr James L

Year: 2012

abstract unavailable at this time

## Data-driven Classification of Screwdriving Operations

Authors: Aronson, Reuben M and Bhatia, Ankit and Jia, Zhenzhong and Guillame-Bert, Mathieu and Bourne, David and Dubrawski, Artur and Mason, Matthew T

Year: 2012

abstract unavailable at this time

## VIPR: An Interactive Tool for Meaningful Visualization of High-Dimensional Data

Authors: Wang, Donghan and Fiterau, Madalina and Dubrawski, Artur

Year: 2012

abstract unavailable at this time

## Explanation-Oriented Classification via Subspace Partitioning

Authors: Fiterau, Madalina and Dubrawski, Artur

Year: 2012

abstract unavailable at this time

## Discovering Compact and Informative Structures through Data Partitioning

Year: 2015

abstract unavailable at this time

## Auton Lab Software Applications Training-of-Trainer Program

Authors: Prashant, Suma and Kanan, T and Waidyanatha, Nuwan and Weerakoon, Pubudini and Gayan, Chinthaka

Year: 2015

abstract unavailable at this time

## ESB\# 331, Indian Institute of Technology Madras (IITM) Chennai--600 036 Website: www. rtbi. in

Authors: Prashant, Suma and Ganesan, M

Year: 2009

abstract unavailable at this time

## Active Transfer Learning

Authors: Wang, Xuezhi

Year: 2016

abstract unavailable at this time

## CMU-ML-15-105 Discovering Compact and Informative Structures through Data Partitioning

Year: 2016

abstract unavailable at this time

## Dynamic and Personalized Risk Forecast in Step-Down Units: Implications for Monitoring Paradigms

Authors: Chen, Lujie and Ogundele, Olufunmilayo and Clermont, Gilles and Hravnak, Marilyn and Pinsky, Michael R and Dubrawski, Artur W

Year: 2016

Venue: Annals of the American Thoracic Society

abstract unavailable at this time

## Fault-Tolerant Source Detection using Bayesian Sensor Reliability Models

Authors: Tandon, P and Huggins, P and Dubrawski, A and Labov, S and Nelson, K

Year: 2016

abstract unavailable at this time

## Project Work Plan

Authors: Waidyanatha, Nuwan

Year: 2016

abstract unavailable at this time

## ASONAM 2012 Program Committee

Authors: Abell, Peter and Ackland, Robert and Adamic, Lada and Afsharchi, Mohsen and Agarwal, Nitin and Akkaya, Kemal and Alani, Harith and Arshad Ali, NIIT and Appel, Pakistan Ana Paula and Arimura, Hiroki and others

Year: 2016

abstract unavailable at this time

## Detection of Radioactive Sources Using Bayesian Aggregation of Data from Mobile Spectrometers

Authors: Tandon, Prateek and Huggins, Peter and Dubrawski, Artur and Labov, Simon and Nelson, Karl

Year: 2016

abstract unavailable at this time

## 2012 IEEE International Conference on Intelligence and Security Informatics

Authors: Cyberspace, Border

Year: 2012

abstract unavailable at this time

## Is there an information hierarchy among hemodynamic variables for early identification of occult hemorrhage?

Authors: Holder, Andre and Guillame-Bert, Mathieu and Chen, Karen and Huggins, Peter and Dubrawski, Artur and Hravnak, Marilyn and Clermont, Gilles and Pinsky, Michael

Year: 2013

Venue: Journal of Critical Care

abstract unavailable at this time

Authors: Kazuo Iwano, IBM and Katz, Japan Randy H and Khatib, Oussama and Truszkowski, Walter F and Verma, Dinesh C and Subhash Wadhva, IIT and Yousif, India Mazim S and de Souza, Jos{\'e} Neuman and Esmahi, Larbi and van der Meer, Sven and others

Year: 2013

abstract unavailable at this time

## Exploiting Non-sequence Data in Dynamic Model Learning

Authors: Huang, Tzu-Kuo

Year: 2013

abstract unavailable at this time

## Rule-Based Anomaly Pattern Detection for Detecting Disease Outbreaks

Authors: Moore, Andrew and Wong, Weng-Keen and Wagner, Mike and Cooper, Greg

Year: 2003

Venue: Proceedings of the 35th Symposium on the Interface of Computing and Statistics

abstract unavailable at this time

## Sub-reviewers

Authors: Parry, Joe and Pavlatos, Christos and Piskorski, Jakub and Pisoiu, Daniela and Probst, Christian and Qu, Guangzhi and Qureshi, Pir Abdul Rasool and Ralevich, Victor and Rhodes, Christopher and Shah, Mahmood H and others

Year: 2003

abstract unavailable at this time

## ICAS Committees

Authors: Kazuo Iwano, IBM and Katz, Japan Randy H and Khatib, Oussama and Truszkowski, Walter F and Verma, Dinesh C and Subhash Wadhva, IIT and Yousif, India Mazim S

Year: 2003

abstract unavailable at this time

## Active Learning for Informative Projection Retrieval

Authors: Data, Labeled Data Unlabeled

Year: 2003

abstract unavailable at this time

## Predictive patterns of sex trafficking online

Authors: Kennedy, Emily

Year: 2012

abstract unavailable at this time

## Robot motion and control 2009

Authors: Koz{\l}owski, Krzysztof

Year: 2009

abstract unavailable at this time

## WATCHING THE CROPS GROW

Authors: Ferber, Dan

Year: 2016

Venue: Mechanical Engineering

abstract unavailable at this time

## Intelligent Robotics Systems—IRS'93

Authors: Crowley, James L

Year: 1994

Venue: Robotics and Autonomous Systems

abstract unavailable at this time

## Infectious disease informatics and biosurveillance

Authors: Zeng, Daniel and Chen, Hsinchun and Bies, Dawn and Lober, William B and Thurmond, Mark

Year: 2010

abstract unavailable at this time

## New perspectives on health information technology

Authors: Harrison, Teresa M

Year: 2010

Venue: Proceedings of the 11th Annual International Digital Government Research Conference on Public Administration Online: Challenges and Opportunities

abstract unavailable at this time

## Suwadana Center Research Assistant Training Workshop

Authors: Center–Kuliyapitiya, Sarvodaya District

Year: 2010

abstract unavailable at this time

## Fleshing things out

Authors: Borror, Connie M

Year: 2012

Venue: Quality Progress

abstract unavailable at this time

## Risk for Cardiorespiratory Instability Following Transfer to a Monitored Step-Down Unit

Authors: Bose, Eliezer and Chen, Lujie and Clermont, Gilles and Dubrawski, Artur and Pinsky, Michael R and Ren, Dianxu and Hoffman, Leslie A and Hravnak, Marilyn

Year: 2017

Venue: Respiratory Care

abstract unavailable at this time

## Session 32AB Leveraging Disruptive Technology to Lead Your Organization From Volume to Value

Authors: Buchler, Richard and Jackson, J Reese and Collens, Steven

Year: 2017

abstract unavailable at this time

## Learning compressible models

Authors: Zhang, Yi and Schneider, Jeff and Dubrawski, Artur

Year: 2010

Venue: Proceedings of the 2010 SIAM International Conference on Data Mining

abstract unavailable at this time

## Environment model adaptation for autonomous exploration

Authors: Nelson, Erik Arthur

Year: 2015

abstract unavailable at this time

## Working Experience

Authors: Lucia, Surname and Indonesian, Nationality

Year: 2014

abstract unavailable at this time

## International Society for Disease Surveillance 10 th Annual Conference 2011

Authors: Neill, Daniel B and Soetebier, Karl A

Year: 2014

abstract unavailable at this time

## Scalable, Flexible and Active Learning on Distributions

Authors: Sutherland, Dougal J

Year: 2016

abstract unavailable at this time

## Ranking-based approaches for localizing faults

Authors: LUCIA, Lucia

Year: 2014

abstract unavailable at this time

## Disjunctive Anomaly Detection: Identifying Complex Anomalous Patterns

Authors: Sabhnani, Robin and Dubrawski, Artur and Schneider, Jeff and Singh, Aarti and Cooper, Gregory

Year: 2010

abstract unavailable at this time

## Intelligence and Security Informatics: Biosurveillance

Authors: Gotham, Daniel Zeng Ivan and Lynch, Ken Komatsu Cecil and Madigan, Mark Thurmond David and Kvach, Bill Lober James and Chen, Hsinchun

Year: 2007

abstract unavailable at this time

## Tractable algorithms for proximity search on large graphs

Authors: Sarkar, Purnamrita

Year: 2010

abstract unavailable at this time

## Modelling Risk of Cardio-Respiratory Instability as a Heterogeneous Process

Authors: Chen, Lujie and Dubrawski, Artur and Clermont, Gilles and Hravnak, Marilyn and Pinsky, Michael R

Year: 2015

Venue: AMIA Annual Symposium Proceedings

abstract unavailable at this time

## Poisson Modeling and Bayesian Estimation of Low Photon Count Signal and Noise Components

Authors: Tandon, Prateek and Huggins, Peter and Dubrawski, Artur and Nelson, Karl and Labov, Simon

Year: 2015

abstract unavailable at this time

## REPORT OF THE TECHNOLOGY TRAINING WORKSHOP FOR VILLAGE HEALTH NURSES/SECTOR HEALTH NURSES, SIVAGANGA DISTRICT, TAMIL NADU, INDIA

Authors: DISTRICT, SIVAGANGA

Year: 2015

abstract unavailable at this time

## Apprentissage de r{\`e}gles associatives temporelles pour les s{\'e}quences temporelles de symboles

Authors: Guillame-Bert, Mathieu

Year: 2012

abstract unavailable at this time

## Canonical Autocorrelation Analysis and Graphical Modeling for Human Trafficking Characterization

Authors: Chen, Qicong and De Arteaga, Maria and Herlands, William

Year: 2012

abstract unavailable at this time

## Detection of Sources of Harmful Radiation using Portable Sensors

Authors: Jin, Jay

Year: 2014

abstract unavailable at this time

## Affordable system for rapid detection and mitigation of emerging diseases

Authors: Gow, Gordon

Year: 2013

Venue: Digital Advances in Medicine, E-Health, and Communication Technologies

abstract unavailable at this time

## Perception of Mobile Phone Data Submission in Real Time Biosurveillance Program by Some Indian Health Workers

Authors: Ganesan, M and Prashant, Suma and Jhunjhunwala, Ashok

Year: 2013

abstract unavailable at this time

## Mobile Robot Localization with an Artificial Neural Network

Authors: Racz, Janusz and Dubrawski, Artur

Year: 1994

Venue: International Workshop on Intelligent Robotic Systems IRS

abstract unavailable at this time

## eHealth Sri lanka 2010

Authors: Dharmaratne, Saminda

Year: 2010

Venue: Sri Lanka Journal of Bio-Medical Informatics

abstract unavailable at this time

## Intelligent robotics systems—SIRS'94

Authors: Crowley, James L

Year: 1995

Venue: Robotics and Autonomous Systems

abstract unavailable at this time

## LASEROWE TR OJWYMIAROWE CZUJNIKI ODLEG LO SCI

Authors: OW, W NAWIGACJI RUCHOMYCH ROBOT

Year: 1995

abstract unavailable at this time

## I would like to express my sincere gratitude to the Conference Committee for their time, dedication and diligence over the past year in delivering an exciting and rewarding scientific conference. We look forward to meeting again next year in Denmark for ASONAM 2010. Best regards, Nicholas Harkiolakis, PhD

Authors: Harkiolakis, Nicholas

Year: 1995

abstract unavailable at this time

## Interpreting Finite Automata for Sequential Data

Authors: Hammerschmidt, Christian Albert and Lin, Qin and Verwer, Sicco and State, Radu

Year: 2016

Venue: arXiv preprint arXiv:1611.07100

abstract unavailable at this time

## SCS Schedule Undergraduate Research Office

Authors: Biswas, Joydeep

Year: 2012

abstract unavailable at this time

## BioCat 2.0

Authors: Corley, Courtney D and Noonan, Christine F and Bartholomew, Rachel A and Franklin, Trisha L and Hutchison, Janine R and Lancaster, Mary J and Madison, Michael C and Piatt, Andrew W

Year: 2013

abstract unavailable at this time

## Multi-agent Planning for Mobile Radiation Source Tracking and Active City-wide Surveillance

Authors: Tandon, Prateek

Year: 2013

abstract unavailable at this time

## CoBi: bio-sensing building mechanical system controls for sustainably enhancing individual thermal comfort

Authors: Choi, Joon Ho

Year: 2010

abstract unavailable at this time

## Learning latent variable and predictive models of dynamical systems

Authors: Siddiqi, Sajid M

Year: 2009

abstract unavailable at this time

## Attentive behavior in an anthropomorphic robot vision system

Authors: Colombo, Carlo and Rucci, Michelle and Dario, Paolo

Year: 1994

Venue: Robotics and Autonomous Systems

abstract unavailable at this time

## Temporal distribution of instability events in continuously monitored step-down unit patients: implications for rapid response systems

Authors: Hravnak, Marilyn and Chen, Lujie and Dubrawski, Artur and Bose, Eliezer and Pinsky, Michael R

Year: 2015

Venue: Resuscitation

abstract unavailable at this time

## Canonical Autocorrelation Analysis for Radiation Threat Detection

Authors: De Arteaga, Maria

Year: 2016

abstract unavailable at this time

## Standard Operating Procedures RealTime Biosurveillance Program

Authors: Waidyanatha, Nuwan

Year: 2009

abstract unavailable at this time

## Real alerts and artifact classification in archived multi-signal vital sign monitoring data: implications for mining big data

Authors: Hravnak, Marilyn and Chen, Lujie and Dubrawski, Artur and Bose, Eliezer and Clermont, Gilles and Pinsky, Michael R

Year: 2016

Venue: Journal of clinical monitoring and computing

abstract unavailable at this time

## Data Analysis Project: $\Sigma$-Optimality for Active Learning on Gaussian Random Fields

Authors: Ma, Yifei

Year: 2016

abstract unavailable at this time

## Detection of radioactive sources in urban scenes using Bayesian Aggregation of data from mobile spectrometers

Authors: Tandon, Prateek and Huggins, Peter and Maclachlan, Rob and Dubrawski, Artur and Nelson, Karl and Labov, Simon

Year: 2016

Venue: Information Systems

abstract unavailable at this time

## Exploring mars using intelligent robots

Authors: Andreou, Paris and Charalambides, Adonis

Year: 1995

Venue: A web-based article, Internet: http://www. doc. ic. ac. uk/\~{} nd/surprise_95/journal/vol4/pma/report. html# Introducti on

abstract unavailable at this time

## Commodification of Flesh: Data Visualization Techniques and Interest in the Licit Sex Industry

Authors: Makin, David A and Bye, Caroline

Year: 2016

Venue: Deviant Behavior

abstract unavailable at this time

## Abouzgheib, Wissam; Coo

Authors: Aagaard, Rasmus and Aakre, Christopher and Ababon, Fides and Abbas, Farrukh and Abbasi, Aleeza and Abd–Allah, Shamel and Abdallah, George and Abdelmalik, Peter and Abdulhadi, Ahmed and Abela, Karla and others

Year: 2016

Venue: Crit Care Med

abstract unavailable at this time

## Evaluation of coded aperture radiation detectors using a Bayesian approach

Authors: Miller, Kyle and Huggins, Peter and Labov, Simon and Nelson, Karl and Dubrawski, Artur

Year: 2016

Venue: Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

abstract unavailable at this time

## Abarquez-Agcaoili

Authors: Abdel-Rasoul, Mahmoud

Year: 2015

Venue: Crit Care Med

abstract unavailable at this time

## Innovation and access to technologies for sustainable development: diagnosing weaknesses and identifying interventions in the transnational arena

Authors: Anadon, Laura Diaz and Matus, Kira JM and Moon, Suerie and Chan, Gabriel and Harley, Alicia and Murthy, Sharmila and Timmer, Vanessa and Latif, Ahmed Abdel and Araujo, Kathleen and Booker, Kayje and others

Year: 2014

abstract unavailable at this time

## Evaluating a Real-Time Biosurveillance Program: A Pilot Project

Authors: Tun, Khan

Year: 2008

abstract unavailable at this time

## Scalable graphical models for social networks

Authors: Goldenberg, Anna

Year: 2007

abstract unavailable at this time

## International Society for Disease Surveillance 10 th Annual Conference 2011 Building the Future of Public Health Surveillance

Authors: Adeoye, Olawunmi and Aman-Oloniyo, Abimbola and Nwaeke, Clement and Mbata, Angela and Oduneye, Abiola and Azarian, Taj and Kite-Powell, Aaron and Zaheer, Saad and Baer, Atar and Kay, Meagan and others

Year: 2007

abstract unavailable at this time

## International Society for Disease Surveillance 10 th Annual Conference 2011 Building the Future of Public Health Surveillance

Authors: Neill, Daniel B and Soetebier, Karl A

Year: 2007

abstract unavailable at this time

## International Society for Disease Surveillance 10 th Annual Conference 2011 Building the Future of Public Health Surveillance

Authors: Neill, Daniel B and Soetebier, Karl A

Year: 2007

abstract unavailable at this time

Authors: Chunara, Rumi and Goetzke, Marie and Brownstein, John

Year: 2007

Venue: International Society for Disease Surveillance 10 th Annual Conference 2011 Building the Future of Public Health Surveillance

abstract unavailable at this time

## Technical Review of the Domestic Nuclear Detection Office Transformational and Applied Research Directorate’s Research and Development Program

Authors: Borchers, Robert and Dahlburg, Jill and Donnelly, John and Isles, Adam and Johnson, Neil and Knoll, Glenn and Kouzes, Richard and Lanza, Richard and Lavietes, Anthony and Lieberman, Jodi and others

Year: 2007

abstract unavailable at this time

## Technical Review of the Domestic Nuclear Detection Office Transformational and Applied Research Directorate's Research and Development Program

Authors: Lavietes, Anthony D and Trebes, James and Borchers, Robert and Dahlburg, Jill and Donnelly, John and Isles, Adam and Johnson, Neil and Knoll, Glenn and Kouzes, Richard and Lanza, Richard and others

Year: 2013

Venue: IEEE Access

abstract unavailable at this time

## The use of mobile phone as a tool for capturing patient data in southern rural Tamil Nadu, India

Authors: MUTHIAH, Ganesan and Prashant, Suma and PUSHPA, Vincy and NATARAJAN, Janakiraman and Jhunjhunwala, Ashok and Waidyanatha, Nuwan

Year: 2011

Venue: Journal of Health Informatics in Developing Countries

abstract unavailable at this time

## Sustainability Science Program

Authors: Araujo, Booker

Year: 2014

abstract unavailable at this time

## Using mobile phones in a real-time biosurveillance program: lessons from the frontlines in Sri Lanka and India

Authors: Gow, Gordon A and Waidyanatha, Nuwan and Mary, Vincy Pushpa

Year: 2010

Venue: Technology and Society (ISTAS), 2010 IEEE International Symposium on

abstract unavailable at this time

## Spatial Data Structures For Efficient Trajectory-Based Queries

Authors: , Jeremy Kubica, Andrew Moore, Andrew Connolly, Robert Jedicke

Year:
Venue: None

Abstract Unavailable At This Time.

Authors: , Jeremy Kubica, Joseph Masiero, Andrew Moore, […], Andrew Connolly

Year:
Venue: None

Abstract Unavailable At This Time.

## T-Cube: Quick Response To Ad-Hoc Time Series Queries Against Large Datasets

Authors: , Maheshkumar Sabhnani, Artur Dubrawski, Andrew Moore

Year:
Venue: None

Abstract Unavailable At This Time.

## Tractable Learning Of Large Bayes Net Structures From Sparse Data

Authors: , Anna Goldenberg, Andrew Moore

Year: Jan 2004

Venue: None

Abstract Unavailable At This Time.

## Fast And Robust Track Initiation Using Multiple Trees

Authors: , Jeremy Kubica, Andrew Moore, Andrew Connolly, Robert Jedicke

Year: Jan 2004

Venue: None

Abstract Unavailable At This Time.

## Wsare: What'S Strange About Recent Events?

Authors: , Weng-Keen Wong, Andrew Moore, Gregory Cooper, Michael Wagner

Year: Jul 2003

Venue: Journal of Urban Health

Abstract Unavailable At This Time.

Authors: , Jeremy Kubica, Andrew Moore, Jeff Schneider, Yiming Yang

Year: Jul 2002

Venue: None

Abstract Unavailable At This Time.

## Data, Network, And Application: Technical Description Of The Utah Rods Winter Olympic Biosurveillance System

Authors: , Fu-Chiang Tsui, Jeremy U Espino, Michael M Wagner, […], Andrew Moore

Year: Feb 2002

Venue: Proceedings / AMIA … Annual Symposium. AMIA Symposium

Abstract Unavailable At This Time.

## Efficient Algorithms For Non-Parametric Clustering With Clutter

Authors: , Weng-keen Wong, Andrew Moore

Year: Jan 2002

Venue: None

Abstract Unavailable At This Time.

## Mix-Nets: Factored Mixtures Of Gaussians In

Authors: , Scott Davies, Andrew Moore

Year: Sep 2001

Venue: None

Abstract Unavailable At This Time.

## Fast Algorithms And Efficient Statistics: N-Point Correlation Functions

Authors: , Andrew Moore, Andy Connolly, Chris Genovese, […], Larry Wasserman

Year: Dec 2000

Venue: None

Abstract Unavailable At This Time.

## Applying Online Search Techniques To Continuous-State Reinforcement Learning

Authors: , Scott Davies, Andrew Y. Ng, Andrew Moore

Year: Sep 2000

Venue: None

Abstract Unavailable At This Time.

## Distributed Value Functions

Authors: , Jeff Schneider, Weng-keen Wong, Andrew Moore, Martin Riedmiller

Year: Jan 2000

Venue: None

Abstract Unavailable At This Time.

## Applying Online Search Techniques To Reinforcement Learning

Authors: , Scott Davies, Andrew Y. Ng, Andrew Moore

Year: Sep 1999

Venue: None

Abstract Unavailable At This Time.

## Bayesian Networks For Lossless Dataset Compression

Authors: , Scott Davies, Andrew Moore

Year: Jul 1999

Venue: None

Abstract Unavailable At This Time.

## Cached Sufficient Statistics For Automated Mining And Discovery From Massive Data Sources

Authors: , Research Showcase, Andrew Moore, Jeff Schneider, […], Paul Komarek

Year: Jan 1999

Venue: None

Abstract Unavailable At This Time.

## Adtrees For Fast Counting And For Fast Learning Of Association Rules

Authors: , Brigham Anderson, Andrew Moore

Year: Oct 1998

Venue: None

Abstract Unavailable At This Time.

## What'S Strange About Recent Events (Wsare) V3.0: Adjusting For A Changing Baseline

Authors: , Weng-Keen Wong, Andrew Moore, Michael Wagner

Year:
Venue: None

Abstract Unavailable At This Time.

## Bayesian Networks For Lossless Compression In Data Mining

Authors: , Scott Davies, Andrew Moore

Year:
Venue: None

Abstract Unavailable At This Time.

## Fast Computation Of The Pair Correlation And N-Point Correlation Functions

Authors: , Alexander Gray, Andrew Moore

Year:
Venue: Computer Physics Communications

Abstract Unavailable At This Time.

## Sequence Selection For Active Learning

Authors: , Brigham Anderson, Sajid Siddiqi, Andrew Moore

Year:
Venue: None

Abstract Unavailable At This Time.

## Logistic Regression For Data Mining And High-Dimensional Classification

Authors: , Paul Komarek, Andrew Moore, Alain Committee, […], Nichol

Year:
Venue: None

Abstract Unavailable At This Time.

## Efficient Analytics For Effective Monitoring Of Biomedical Security

Authors: , Maheshkumar Sabhnani, Daniel Neill, Andrew Moore, […], Weng-Keen Wong

Year:
Venue: None

Abstract Unavailable At This Time.

## Mix-Nets: Factored Mixtures Of Gaussians In Bayesian Networks With Mixed

Continuous And Discrete Variables

Authors: , Scott Davies, Andrew Moore

Year: Jan 2013

Venue: None

Abstract Unavailable At This Time.

## Interpolating Conditional Density Trees

Authors: , Scott Davies, Andrew Moore

Year: Dec 2012

Venue: None

Abstract Unavailable At This Time.

## Efficient Intra- And Inter-Night Linking Of Asteroid Detections Using Kd-Trees

Authors: , Jeremy Kubica, Larry Denneau, Tommy Grav, […], Richard J. Wainscoat

Year: Mar 2007

Venue: Icarus

Abstract Unavailable At This Time.

## Scalable Detection And Optimization Of N-Ary Linkages

Authors: , Andrew Moore, Jeff Schneider, Jeremy Kubica, […], Purna Sarkar

Year: Jun 2006

Venue: None

Abstract Unavailable At This Time.

## Nonparametric Bayesian Classification With Massive Datasets: Large-Scale Quasar Discovery

Authors: , ALEXANDER GRAY, GORDON RICHARDS, ROBERT NICHOL, […], ANDREW MOORE

Year: May 2006

Venue: None

Abstract Unavailable At This Time.

## Implementation Of Logistic Regression With Truncated Irls

Authors: , Paul Komarek, Andrew Moore

Year: Jan 2006

Venue: None

Abstract Unavailable At This Time.

## Massive Science With Vo And Grids

Authors: , Robert Nichol, Garry Smith, Christopher Miller, […], Andrew Moore

Year: Dec 2005

Venue: None

Abstract Unavailable At This Time.

## Efficiently Identifying Close Track/Observation Pairs In Continuous Timed Data - Art. No. 59130S

Authors: , Jeremy Kubica, Andrew Moore, Andrew Connolly, Robert Jedicke

Year: Sep 2005

Venue: Proceedings of SPIE - The International Society for Optical Engineering

Abstract Unavailable At This Time.

## Making Logistic Regression A Core Data Mining Tool A Practical Investigation Of Accuracy, Speed, And Simplicity

Authors: , Paul Komarek, Andrew Moore

Year: Apr 2005

Venue: None

Abstract Unavailable At This Time.

## Active Learning For Hidden Markov Models: Objective Functions And Algorithms.

Authors: , Brigham Anderson, Andrew Moore

Year: Jan 2005

Venue: None

Abstract Unavailable At This Time.

## Efficient Algorithms For The Identification Of Potential Track/Observation Associations In Continuous Time Data

Authors: , Jeremy Kubica, Andrew Moore, Andrew Connolly, Robert Jedicke

Year: Jan 2005

Venue: None

Abstract Unavailable At This Time.

## Rule-Based Anomaly Pattern Detection For Detecting Disease Outbreaks

Authors: , Weng-keen Wong, Andrew Moore, Gregory Cooper, Michael Wagner

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

Authors: , Anna Goldenberg, Andrew Moore

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

## Unknown

Authors: , Weng-keen Wong, Andrew Moore, Gregory Cooper, Michael Wagner

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

Authors: , Jeremy Kubica, Andrew Moore, David Cohn, Jeff Schneider

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

## Finding Underlying Connections:

Authors: , Jeremy Kubica, Andrew Moore, David Cohn, Jeff Schneider

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

## Probabilistic Noise Identification And Data Cleaning

Authors: , Jeremy Kubica, Andrew Moore

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

## Optimal Reinsertion: A New Search Operator For Accelerated And More Accurate Bayesian Network Structure Learning

Authors: , Andrew Moore, Weng-keen Wong

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

Authors: , Paul Komarek, Andrew Moore

Year: May 2004

Venue: None

Abstract Unavailable At This Time.

## Ecient Algorithms For Non-Parametric

Authors: , Weng-keen Wong, Andrew Moore

Year: May 2004

Venue: None

Abstract Unavailable At This Time.