Maximum Entropy Correlated Equilibria

We study maximum entropy correlated equilibria in (multi-player)games and provide two gradient-based algorithms that are guaranteedto converge to such equilibria. Although we do not provideconvergence rates for these algorithms, they do have strong connectionsto other algorithms (such as iterative s...

Full description

Bibliographic Details
Main Authors: Ortiz, Luis E., Schapire, Robert E., Kakade, Sham M.
Other Authors: Leslie Kaelbling
Language:en_US
Published: 2006
Subjects:
Online Access:http://hdl.handle.net/1721.1/31339
_version_ 1811085595764588544
author Ortiz, Luis E.
Schapire, Robert E.
Kakade, Sham M.
author2 Leslie Kaelbling
author_facet Leslie Kaelbling
Ortiz, Luis E.
Schapire, Robert E.
Kakade, Sham M.
author_sort Ortiz, Luis E.
collection MIT
description We study maximum entropy correlated equilibria in (multi-player)games and provide two gradient-based algorithms that are guaranteedto converge to such equilibria. Although we do not provideconvergence rates for these algorithms, they do have strong connectionsto other algorithms (such as iterative scaling) which are effectiveheuristics for tasks such as statistical estimation.
first_indexed 2024-09-23T13:12:06Z
id mit-1721.1/31339
institution Massachusetts Institute of Technology
language en_US
last_indexed 2024-09-23T13:12:06Z
publishDate 2006
record_format dspace
spelling mit-1721.1/313392019-04-11T06:23:39Z Maximum Entropy Correlated Equilibria Ortiz, Luis E. Schapire, Robert E. Kakade, Sham M. Leslie Kaelbling Learning and Intelligent Systems Game Theory, Graphical Games, Graphical Models, Normal-form Games, Optimization We study maximum entropy correlated equilibria in (multi-player)games and provide two gradient-based algorithms that are guaranteedto converge to such equilibria. Although we do not provideconvergence rates for these algorithms, they do have strong connectionsto other algorithms (such as iterative scaling) which are effectiveheuristics for tasks such as statistical estimation. 2006-03-20T19:24:59Z 2006-03-20T19:24:59Z 2006-03-20 MIT-CSAIL-TR-2006-021 http://hdl.handle.net/1721.1/31339 en_US Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 15 p. 301190 bytes 876790 bytes application/pdf application/postscript application/pdf application/postscript
spellingShingle Game Theory, Graphical Games, Graphical Models, Normal-form Games, Optimization
Ortiz, Luis E.
Schapire, Robert E.
Kakade, Sham M.
Maximum Entropy Correlated Equilibria
title Maximum Entropy Correlated Equilibria
title_full Maximum Entropy Correlated Equilibria
title_fullStr Maximum Entropy Correlated Equilibria
title_full_unstemmed Maximum Entropy Correlated Equilibria
title_short Maximum Entropy Correlated Equilibria
title_sort maximum entropy correlated equilibria
topic Game Theory, Graphical Games, Graphical Models, Normal-form Games, Optimization
url http://hdl.handle.net/1721.1/31339
work_keys_str_mv AT ortizluise maximumentropycorrelatedequilibria
AT schapireroberte maximumentropycorrelatedequilibria
AT kakadeshamm maximumentropycorrelatedequilibria