2026-07-24T00:57:02Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:12833 2024-03-06T10:50:48Z ddc:004 open_access

Model-Free Reinforcement Learning for Stochastic Parity Games Hahn, Ernst Moritz Perez, Mateo Schewe, Sven Somenzi, Fabio Trivedi, Ashutosh Wojtczak, Dominik Reinforcement learning Stochastic games Omega-regular objectives This paper investigates the use of model-free reinforcement learning to compute the optimal value in two-player stochastic games with parity objectives. In this setting, two decision makers, player Min and player Max, compete on a finite game arena - a stochastic game graph with unknown but fixed probability distributions - to minimize and maximize, respectively, the probability of satisfying a parity objective. We give a reduction from stochastic parity games to a family of stochastic reachability games with a parameter ε, such that the value of a stochastic parity game equals the limit of the values of the corresponding simple stochastic games as the parameter ε tends to 0. Since this reduction does not require the knowledge of the probabilistic transition structure of the underlying game arena, model-free reinforcement learning algorithms, such as minimax Q-learning, can be used to approximate the value and mutual best-response strategies for both players in the underlying stochastic parity game. We also present a streamlined reduction from 1 1/2-player parity games to reachability games that avoids recourse to nondeterminism. Finally, we report on the experimental evaluations of both reductions. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Ernst Moritz Hahn and Mateo Perez and Sven Schewe and Fabio Somenzi and Ashutosh Trivedi and Dominik Wojtczak 2020 Is Part Of LIPIcs, Volume 171, 31st International Conference on Concurrency Theory (CONCUR 2020) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.CONCUR.2020.21 urn:nbn:de:0030-drops-128332 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.CONCUR.2020.21 eng https://creativecommons.org/licenses/by/3.0/legalcode