English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
 Local TagsRelease HistoryDetailsSummary
  Reinforcement Learning with Bounded Information Loss

Peters, J., Mülling, K., Seldin, Y., & Altun, Y. (2011). Reinforcement Learning with Bounded Information Loss. In A. Mohammad-Djafari, J.-F. Bercher, & P. Bessière (Eds.), AIP Conference Proceedings (pp. 365-372). Woodbury, NY, USA: American Institute of Physics. doi:10.1063/1.3573639.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/21.11116/0000-0002-81C9-9 Version Permalink: https://hdl.handle.net/21.11116/0000-0002-81CA-8
Genre: Conference Paper

Files

show Files

Locators

show
hide
Locator:
https://aip.scitation.org/doi/pdf/10.1063/1.3573639?class=pdf (Publisher version)
Description:
-
OA-Status:

Creators

show
hide
 Creators:
Peters, J1, 2, Author           
Mülling, K1, 2, Author           
Seldin, Y1, 2, Author           
Altun, Y1, 2, Author           
Affiliations:
1Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795              
2Max Planck Institute for Biological Cybernetics, Max Planck Society, Spemannstrasse 38, 72076 Tübingen, DE, ou_1497794              

Content

show
hide
Free keywords: -
 Abstract: Policy search is a successful approach to reinforcement learning. However, policy improvements often result in the loss of information. Hence, it has been marred by premature convergence and implausible solutions. As first suggested in the context of covariant or natural policy gradients, many of these problems may be addressed by constraining the information loss. In this paper, we continue this path of reasoning and suggest two reinforcement learning methods, i.e., a model‐based and a model free algorithm that bound the loss in relative entropy while maximizing their return. The resulting methods differ significantly from previous policy gradient approaches and yields an exact update step. It works well on typical reinforcement learning benchmark problems as well as novel evaluations in robotics. We also show a Bayesian bound motivation of this new approach [8].

Details

show
hide
Language(s):
 Dates: 2011-03
 Publication Status: Published online
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: DOI: 10.1063/1.3573639
 Degree: -

Event

show
hide
Title: 30th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering (MaxEnt 2010)
Place of Event: Chamonix, France
Start-/End Date: 2010-07-04 - 2010-07-09

Legal Case

show

Project information

show

Source 1

show
hide
Title: AIP Conference Proceedings
Source Genre: Proceedings
 Creator(s):
Mohammad-Djafari, A, Editor
Bercher, J-F, Editor
Bessière, P, Editor
Affiliations:
-
Publ. Info: Woodbury, NY, USA : American Institute of Physics
Pages: - Volume / Issue: 1305 (1) Sequence Number: - Start / End Page: 365 - 372 Identifier: ISBN: 978-073540860-9