A-PSRO: A Unified Strategy Learning Method with Advantage Function for Normal-form Games

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Hu, Yudong, Li, Haoran, Han, Congying, Guo, Tiande, Li, Mingqiang, Li, Bonan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916171434950656
author Hu, Yudong
Li, Haoran
Han, Congying
Guo, Tiande
Li, Mingqiang
Li, Bonan
author_facet Hu, Yudong
Li, Haoran
Han, Congying
Guo, Tiande
Li, Mingqiang
Li, Bonan
contents Solving Nash equilibrium is the key challenge in normal-form games with large strategy spaces, where open-ended learning frameworks offer an efficient approach. In this work, we propose an innovative unified open-ended learning framework A-PSRO, i.e., Advantage Policy Space Response Oracle, as a comprehensive framework for both zero-sum and general-sum games. In particular, we introduce the advantage function as an enhanced evaluation metric for strategies, enabling a unified learning objective for agents engaged in normal-form games. We prove that the advantage function exhibits favorable properties and is connected with the Nash equilibrium, which can be used as an objective to guide agents to learn strategies efficiently. Our experiments reveal that A-PSRO achieves a considerable decrease in exploitability in zero-sum games and an escalation in rewards in general-sum games, significantly outperforming previous PSRO algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2308_12520
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A-PSRO: A Unified Strategy Learning Method with Advantage Function for Normal-form Games
Hu, Yudong
Li, Haoran
Han, Congying
Guo, Tiande
Li, Mingqiang
Li, Bonan
Computer Science and Game Theory
Solving Nash equilibrium is the key challenge in normal-form games with large strategy spaces, where open-ended learning frameworks offer an efficient approach. In this work, we propose an innovative unified open-ended learning framework A-PSRO, i.e., Advantage Policy Space Response Oracle, as a comprehensive framework for both zero-sum and general-sum games. In particular, we introduce the advantage function as an enhanced evaluation metric for strategies, enabling a unified learning objective for agents engaged in normal-form games. We prove that the advantage function exhibits favorable properties and is connected with the Nash equilibrium, which can be used as an objective to guide agents to learn strategies efficiently. Our experiments reveal that A-PSRO achieves a considerable decrease in exploitability in zero-sum games and an escalation in rewards in general-sum games, significantly outperforming previous PSRO algorithms.
title A-PSRO: A Unified Strategy Learning Method with Advantage Function for Normal-form Games
topic Computer Science and Game Theory
url https://arxiv.org/abs/2308.12520