Kernel-Based Reinforcement Learning

Kernel-Based Reinforcement Learning (1999)

Cached

Download Links

[www.jax.org]

Other Repositories/Bibliography

DBLP

by Dirk Ormoneit , Saunak Sen

Venue:	Machine Learning
Citations:	79 - 1 self

Versions

Version History

Metadata Version 3

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	Kernel-Based Reinforcement Learning	user correction - Legacy Corrections
AUTHOR NAME	Dirk Ormoneit	SVM HeaderParse 0.1
AUTHOR AFFIL	Department of Statistics, Stanford University, Stanford	SVM HeaderParse 0.2
AUTHOR ADDR	CA 94305-4065	SVM HeaderParse 0.1
AUTHOR NAME	Saunak Sen	SVM HeaderParse 0.1
AUTHOR AFFIL	Department of Statistics, Stanford University, Stanford	SVM HeaderParse 0.2
AUTHOR ADDR	CA 94305-4065	SVM HeaderParse 0.1
ABSTRACT	We present a kernel-based approach to reinforcement learning that overcomes the stability problems of temporal-difference learning in continuous state-spaces. First, our algorithm converges to a unique solution of an approximate Bellman's equation regardless of its initialization values. Second, the method is consistent in the sense that the resulting policy converges asymptotically to the optimal policy. Parametric value function estimates such as neural networks do not possess this property. Our kernel-based approach also allows us to show that the limiting distribution of the value function estimate is a Gaussian process. This information is useful in studying the bias-variance tradeo in reinforcement learning. We find that all reinforcement learning approaches to estimating the value function, parametric or non-parametric, are subject to a bias. This bias is typically larger in reinforcement learning than in a comparable regression problem.	user correction - Legacy Corrections
YEAR	1999	user correction - Legacy Corrections
VENUE	Machine Learning	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
PAGES	161--178	INFERENCE
VOLUME	49	INFERENCE
CITATIONS	35 found	ParsCit 1.0