Experts in a markov decision process

Eyal Even-Dar, Sham M. Kakade, Yishay Mansour

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We consider an MDP setting in which the reward function is allowed to change during each time step of play (possibly in an adversarial manner), yet the dynamics remain fixed. Similar to the experts setting, we address the question of how well can an agent do when compared to the reward achieved under the best stationary policy over time. We provide efficient algorithms, which have regret bounds with no dependence on the size of state space. Instead, these bounds depend only on a certain horizon time of the process and logarithmically on the number of actions. We also show that in the case that the dynamics change over time, the problem becomes computationally hard.

Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004
PublisherNeural information processing systems foundation
ISBN (Print)0262195348, 9780262195348
StatePublished - 2005
Event18th Annual Conference on Neural Information Processing Systems, NIPS 2004 - Vancouver, BC, Canada
Duration: 13 Dec 200416 Dec 2004

Publication series

NameAdvances in Neural Information Processing Systems
ISSN (Print)1049-5258

Conference

Conference18th Annual Conference on Neural Information Processing Systems, NIPS 2004
Country/TerritoryCanada
CityVancouver, BC
Period13/12/0416/12/04

Fingerprint

Dive into the research topics of 'Experts in a markov decision process'. Together they form a unique fingerprint.

Cite this