English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
Add to Basket
 Local TagsRelease HistoryDetailsSummary
  Coherent Inference on Optimal Play in Game Trees

Hennig, P., Stern, D., & Graepel, T. (2010). Coherent Inference on Optimal Play in Game Trees. In Y. Teh, & M. Titterington (Eds.), JMLR Workshop and Conference Proceedings (pp. 326-333). Cambridge, MA, USA: JMLR.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/11858/00-001M-0000-0013-C03C-1 Version Permalink: https://hdl.handle.net/21.11116/0000-0002-81EF-F
Genre: Conference Paper

Files

show Files

Locators

show
hide
Locator:
http://proceedings.mlr.press/v9/hennig10a/hennig10a.pdf (Publisher version)
Description:
-
OA-Status:

Creators

show
hide
 Creators:
Hennig, P1, Author           
Stern, D, Author
Graepel, T, Author
Affiliations:
1Cavendish Laboratory, Cambridge UK, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Round-based games are an instance of discrete planning problems. Some of the best contemporary game tree search algorithms use random roll-outs as data. Relying on a good policy, they learn on-policy values by propagating information upwards in the tree, but not between sibling nodes. Here, we present a generative model and a corresponding approximate message passing scheme for inference on the optimal, off-policy value of nodes in smooth AND/OR trees, given random roll-outs. The crucial insight is that the distribution of values in game trees is not completely arbitrary. We define a generative model of the on-policy values using a latent score for each state, representing the value under the random roll-out policy. Inference on the values under the optimal policy separates into an inductive, pre-data step and a deductive, post-data part. Both can be solved approximately with Expectation Propagation, allowing off-policy value inference for any node in the (exponentially big) tree in linear time.

Details

show
hide
Language(s):
 Dates: 2010-05
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: BibTex Citekey: HennigSG2010
 Degree: -

Event

show
hide
Title: Thirteenth International Conference on Artificial Intelligence and Statistics (AI & Statistics 2010)
Place of Event: Chia Laguna Resort, Italy
Start-/End Date: 2010-05-13 - 2010-05-15

Legal Case

show

Project information

show

Source 1

show
hide
Title: JMLR Workshop and Conference Proceedings
Source Genre: Proceedings
 Creator(s):
Teh, YW, Editor
Titterington, M, Editor
Affiliations:
-
Publ. Info: Cambridge, MA, USA : JMLR
Pages: - Volume / Issue: 9 Sequence Number: - Start / End Page: 326 - 333 Identifier: -