Computing Maxmin Strategies in Extensive-form Zero-sum Games with Imperfect Recall

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F17%3A00315209" target="_blank" >RIV/68407700:21230/17:00315209 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.5220/0006121200630074" target="_blank" >http://dx.doi.org/10.5220/0006121200630074</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.5220/0006121200630074" target="_blank" >10.5220/0006121200630074</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Computing Maxmin Strategies in Extensive-form Zero-sum Games with Imperfect Recall

Popis výsledku v původním jazyce

Extensive-form games with imperfect recall are an important game-theoretic model that allows a compact representation of strategies in dynamic strategic interactions. Practical use of imperfect recall games is limited due to negative theoretical results: a Nash equilibrium does not have to exist, computing maxmin strategies is NP-hard, and they may require irrational numbers. We present the first algorithm for approximating maxmin strategies in two-player zero-sum imperfect recall games without absentmindedness. We modify the wellknown sequence-form linear program to model strategies in imperfect recall games resulting in a bilinear program and use a recent technique to approximate the bilinear terms. Our main algorithm is a branch-andbound search that provably reaches the desired approximation after an exponential number of steps in the size of the game. Experimental evaluation shows that the proposed algorithm can approximate maxmin strategies of randomly generated imperfect recall games of sizes beyond toy-problems within few minutes.

Název v anglickém jazyce

Computing Maxmin Strategies in Extensive-form Zero-sum Games with Imperfect Recall

Popis výsledku anglicky

Extensive-form games with imperfect recall are an important game-theoretic model that allows a compact representation of strategies in dynamic strategic interactions. Practical use of imperfect recall games is limited due to negative theoretical results: a Nash equilibrium does not have to exist, computing maxmin strategies is NP-hard, and they may require irrational numbers. We present the first algorithm for approximating maxmin strategies in two-player zero-sum imperfect recall games without absentmindedness. We modify the wellknown sequence-form linear program to model strategies in imperfect recall games resulting in a bilinear program and use a recent technique to approximate the bilinear terms. Our main algorithm is a branch-andbound search that provably reaches the desired approximation after an exponential number of steps in the size of the game. Experimental evaluation shows that the proposed algorithm can approximate maxmin strategies of randomly generated imperfect recall games of sizes beyond toy-problems within few minutes.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA15-23235S" target="_blank" >GA15-23235S: Abstrakce a extenzivní hry s nedokonalou pamětí</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 9th International Conference on Agents and Artificial Intelligence - Volume 2

ISBN

978-989-758-220-2

ISSN

—

e-ISSN

—

Počet stran výsledku

12

Strana od-do

63-74

Název nakladatele

SciTePress - Science and Technology Publications

Místo vydání

Porto

Místo konání akce

Porto

Datum konání akce

24. 2. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000413244200006