Yi Zhang
Graducate student
Biography
I am a PhD student at Machine Learning Department, School of Computer Science, Carnegie Mellon University. I work with Jeff Schneider on learning with limited supervision by encoding extra side information. I'm also interested in large-scale web mining and parallel machine learning (see my website for details). I've worked as summer interns on behavioral targeting and computational advertising at Yahoo! Labs and parallel machine learning on Hadoop at IBM T.J. Watson.
My research has been supported by IBM PhD Fellowships and Yahoo! Key Scientific Challenges Awards. Thanks IBM and Yahoo!
Papers
-
A Composite Likelihood View for Multi-Label Classification
(2012)
A Composite Likelihood View for Multi-Label Classification -
Maximum Margin Output Coding
(2012)
Maximum Margin Output Coding -
Multi-label Output Codes using Canonical Correlation Analysis
(2011)
Multi-label Output Codes using Canonical Correlation Analysis -
Learning Multiple Tasks with a Sparse Matrix-Normal Penalty
(2010)
-
Multi-task Active Learning with Output Constraints
(2010)
A multi-task active learning framework with task outputs coupled by constraints -
Projection Penalties: Dimension Reduction without Loss
(2010)
Projection Penalties for Using Dimension Reduction without Loss -
Learning Compressible Models
(2010)
Learning Compressible Models -
Smart PCA
(2009)
Smart PCA algorithm to incorporate domain knowledge into dimension reduction -
Learning the Semantic Correlation: An Alternative Way to Gain from Unlabeled Text
(2008)
A semi-supervised learning algorithm on text
Software
-
Source Code: Multi-Label Output Codes using Canonical Correlation Analysis
Multi-label Output Code using Canonical Correlation Analysis -
Source Code: Maximum Margin Output Coding
Multi-label Output Code using Canonical Correlation Analysis