Learning not to regret

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F24%3A00378617" target="_blank" >RIV/68407700:21230/24:00378617 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/00216208:11320/24:10490822

Výsledek na webu

<a href="https://doi.org/10.1609/aaai.v38i14.29443" target="_blank" >https://doi.org/10.1609/aaai.v38i14.29443</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1609/aaai.v38i14.29443" target="_blank" >10.1609/aaai.v38i14.29443</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Learning not to regret

Popis výsledku v původním jazyce

The literature on game-theoretic equilibrium finding predominantly focuses on single games or their repeated play. Nevertheless, numerous real-world scenarios feature playing a game sampled from a distribution of similar, but not identical games, such as playing poker with different public cards or trading correlated assets on the stock market. As these similar games feature similar equilibra, we investigate a way to accelerate equilibrium finding on such a distribution. We present a novel "learning not to regret" framework, enabling us to meta-learn a regret minimizer tailored to a specific distribution. Our key contribution, Neural Predictive Regret Matching, is uniquely meta-learned to converge rapidly for the chosen distribution of games, while having regret minimization guarantees on any game. We validated our algorithms' faster convergence on a distribution of river poker games. Our experiments show that the meta-learned algorithms outpace their non-meta-learned counterparts, achieving more than tenfold improvements.

Název v anglickém jazyce

Learning not to regret

Popis výsledku anglicky

The literature on game-theoretic equilibrium finding predominantly focuses on single games or their repeated play. Nevertheless, numerous real-world scenarios feature playing a game sampled from a distribution of similar, but not identical games, such as playing poker with different public cards or trading correlated assets on the stock market. As these similar games feature similar equilibra, we investigate a way to accelerate equilibrium finding on such a distribution. We present a novel "learning not to regret" framework, enabling us to meta-learn a regret minimizer tailored to a specific distribution. Our key contribution, Neural Predictive Regret Matching, is uniquely meta-learned to converge rapidly for the chosen distribution of games, while having regret minimization guarantees on any game. We validated our algorithms' faster convergence on a distribution of river poker games. Our experiments show that the meta-learned algorithms outpace their non-meta-learned counterparts, achieving more than tenfold improvements.

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/GA22-26655S" target="_blank" >GA22-26655S: Algoritmy pro hraní masivních her s neúplnou informací</a><br>

Návaznosti

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

Rok uplatnění

2024

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the 38th AAAI Conference on Artificial Intelligence

ISBN

978-1-57735-887-9

ISSN

2159-5399

e-ISSN

2374-3468

Počet stran výsledku

9

Strana od-do

15202-15210

Název nakladatele

AAAI Press

Místo vydání

Menlo Park

Místo konání akce

Vancouver

Datum konání akce

20. 2. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001239983500003