Bayesian Reinforcement Learning

Nikos Vlassis 1 Mohammad Ghavamzadeh 2 Shie Mannor 3 Pascal Poupart 4
2 SEQUEL - Sequential Learning
LIFL - Laboratoire d'Informatique Fondamentale de Lille, Inria Lille - Nord Europe, LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
Abstract : This chapter surveys recent lines of work that use Bayesian techniques for reinforcement learning. In Bayesian learning, uncertainty is expressed by a prior distribution over unknown parameters and learning is achieved by computing a posterior distribution based on the data observed. Hence, Bayesian reinforcement learning distinguishes itself from other forms of reinforcement learning by explicitly maintaining a distribution over various quantities such as the parameters of the model, the value function, the policy or its gradient. This yields several benefits: a) domain knowledge can be naturally encoded in the prior distribution to speed up learning; b) the exploration/exploitation tradeoff can be naturally optimized; and c) notions of risk can be naturally taken into account to obtain robust policies.
Document type :
Book sections
Complete list of metadatas

https://hal.inria.fr/hal-00840479
Contributor : Mohammad Ghavamzadeh <>
Submitted on : Tuesday, July 2, 2013 - 3:31:33 PM
Last modification on : Thursday, February 21, 2019 - 10:52:49 AM
Long-term archiving on : Thursday, October 3, 2013 - 10:45:23 AM

File

BRLchapter.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00840479, version 1

Collections

Citation

Nikos Vlassis, Mohammad Ghavamzadeh, Shie Mannor, Pascal Poupart. Bayesian Reinforcement Learning. Marco Wiering and Martijn van Otterlo. Reinforcement Learning: State of the Art, Springer Verlag, 2012. ⟨hal-00840479⟩

Share

Metrics

Record views

714

Files downloads

2235