Mastering the game of Stratego with model-free multiagent reinforcement learning

被引：48

作者：

Perolat, Julien ^{[1
]}

De Vylder, Bart ^{[1
]}

Hennes, Daniel ^{[1
]}

Tarassov, Eugene ^{[1
]}

Strub, Florian ^{[1
]}

de Boer, Vincent ^{[1
]}

Muller, Paul ^{[1
]}

Connor, Jerome T. ^{[1
]}

Burch, Neil ^{[1
]}

Anthony, Thomas ^{[1
]}

McAleer, Stephen ^{[1
]}

Elie, Romuald ^{[1
]}

Cen, Sarah H. ^{[1
]}

Wang, Zhe ^{[1
]}

Gruslys, Audrunas ^{[1
]}

Malysheva, Aleksandra ^{[1
]}

Khan, Mina ^{[1
]}

Ozair, Sherjil ^{[1
]}

Timbers, Finbarr ^{[1
]}

Pohlen, Toby ^{[1
]}

Eccles, Tom ^{[1
]}

Rowland, Mark ^{[1
]}

Lanctot, Marc ^{[1
]}

Lespiau, Jean-Baptiste ^{[1
]}

Piot, Bilal ^{[1
]}

Omidshafiei, Shayegan ^{[1
]}

Lockhart, Edward ^{[1
]}

Sifre, Laurent ^{[1
]}

Beauguerlange, Nathalie ^{[1
]}

Munos, Remi ^{[1
]}

Silver, David ^{[1
]}

Singh, Satinder ^{[1
]}

Hassabis, Demis ^{[1
]}

Tuyls, Karl ^{[1
]}

机构：

[1] DeepMind Technol Ltd, London, England

来源：

SCIENCE | 2022年 / 378卷 / 6623期

关键词：

LEVEL; DYNAMICS; GO;

D O I：

10.1126/science.add4679

中图分类号：

O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

We introduce DeepNash, an autonomous agent that plays the imperfect information game Stratego at a human expert level. Stratego is one of the few iconic board games that artificial intelligence (AI) has not yet mastered. It is a game characterized by a twin challenge: It requires long-term strategic thinking as in chess, but it also requires dealing with imperfect information as in poker. The technique underpinning DeepNash uses a game-theoretic, model-free deep reinforcement learning method, without search, that learns to master Stratego through self-play from scratch. DeepNash beat existing stateof-the-art AI methods in Stratego and achieved a year-to-date (2022) and all-time top-three ranking on the Gravon games platform, competing with human expert players.

引用

页码：990 / +

页数：7

共 50 条

[1] Model-Free Reinforcement Learning for Fully Cooperative Consensus Problem of Nonlinear Multiagent Systems
Wang, Hong
Li, Man
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 33 (04) : 1482 - 1491
[2] Model-Free Nonstationary Reinforcement Learning: Near-Optimal Regret and Applications in Multiagent Reinforcement Learning and Inventory Control
Mao, Weichao
Zhang, Kaiqing
Zhu, Ruihao
Simchi-Levi, David
Basar, Tamer
[J]. MANAGEMENT SCIENCE, 2024,
[3] Learning Representations in Model-Free Hierarchical Reinforcement Learning
Rafati, Jacob
Noelle, David C.
[J]. THIRTY-THIRD AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FIRST INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / NINTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2019, : 10009 - 10010
[4] Model-Free Trajectory Optimization for Reinforcement Learning
Akrour, Riad
Abdolmaleki, Abbas
Abdulsamad, Hany
Neumann, Gerhard
[J]. INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 48, 2016, 48
[5] Model-Free Quantum Control with Reinforcement Learning
Sivak, V. V.
Eickbusch, A.
Liu, H.
Royer, B.
Tsioutsios, I
Devoret, M. H.
[J]. PHYSICAL REVIEW X, 2022, 12 (01)
[6] Model-Free Active Exploration in Reinforcement Learning
Russo, Alessio
Proutiere, Alexandre
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
[7] Online Nonstochastic Model-Free Reinforcement Learning
Ghai, Udaya
Gupta, Arushi
Xia, Wenhan
Singh, Karan
Hazan, Elad
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
[8] Recovering Robustness in Model-Free Reinforcement Learning
Venkataraman, Harish K.
Seiler, Peter J.
[J]. 2019 AMERICAN CONTROL CONFERENCE (ACC), 2019, : 4210 - 4216
[9] Model-Free Reinforcement Learning Algorithms: A Survey
Calisir, Sinan
Pehlivanoglu, Meltem Kurt
[J]. 2019 27TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2019,
[10] Model-Free Event-Triggered Consensus Algorithm for Multiagent Systems Using Reinforcement Learning Method
Long, Mingkang
Su, Housheng
Zeng, Zhigang
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2022, 52 (08): : 5212 - 5221

← 1 2 3 4 5 →