Prioritized Sweeping: Reinforcement Learning with Less Data and Less Real Time (1993)
Tags
Markov Decision Processes, Reinforcement Learning
Abstract
We present a new algorithm, Prioritized Sweeping, for efficient prediction and control of stochastic Markov systems. Incremental learning methods such as Temporal Differencing and Qlearning have fast real time performance. Classical methods are slower, but more accurate, because they make full use of the observations. Prioritized Sweeping aims for the best of both worlds. It uses all previous experiences both to prioritize important dynamic programming sweeps and to guide the exploration of state-space. We compare Prioritized Sweeping with other reinforcement learning schemes for a number of different stochastic optimal control problems. It successfully solves large state-space real time problems with which other methods have difficulty.
Full text
Download (application/pdf, 621.9 kB)
Approximate BibTeX Entry
@article{moore-prioritized,
Year = {1993},
Journal = {Machine Learning},
Volume = {13},
Pages = {103-130},
Publisher = {Kluwer Academic Publishers},
Address = {Boston},
Author = {
Andrew Moore, Chris
Atkeson
},
Title = {Prioritized Sweeping: Reinforcement Learning with Less Data and Less Real Time}
}