Reinforcement Learning in Multi-agent Games: Open AI Gym Diplomacy Environment
D. Cruz, J. Cruz, and H. Lopes Cardoso. Progress in Artificial Intelligence, page 49--60. Cham, Springer International Publishing, (2019)
Abstract
Reinforcement learning has been successfully applied to adversarial games, exhibiting its potential. However, most real-life scenarios also involve cooperation, in addition to competition. Using reinforcement learning in multi-agent cooperative games is, however, still mostly unexplored. In this paper, a reinforcement learning environment for the Diplomacy board game is presented, using the standard interface adopted by OpenAI Gym environments. Our main purpose is to enable straightforward comparison and reuse of existing reinforcement learning implementations when applied to cooperative games. As a proof-of-concept, we show preliminary results of reinforcement learning agents exploiting this environment.
Description
Reinforcement Learning in Multi-agent Games: Open AI Gym Diplomacy Environment | SpringerLink
%0 Conference Paper
%1 cruz2019diplomacy
%A Cruz, Diogo
%A Cruz, José Aleixo
%A Lopes Cardoso, Henrique
%B Progress in Artificial Intelligence
%C Cham
%D 2019
%E Moura Oliveira, Paulo
%E Novais, Paulo
%E Reis, Luís Paulo
%I Springer International Publishing
%K ACKTR PPO environment multi_agent reinforcement_learning
%P 49--60
%T Reinforcement Learning in Multi-agent Games: Open AI Gym Diplomacy Environment
%X Reinforcement learning has been successfully applied to adversarial games, exhibiting its potential. However, most real-life scenarios also involve cooperation, in addition to competition. Using reinforcement learning in multi-agent cooperative games is, however, still mostly unexplored. In this paper, a reinforcement learning environment for the Diplomacy board game is presented, using the standard interface adopted by OpenAI Gym environments. Our main purpose is to enable straightforward comparison and reuse of existing reinforcement learning implementations when applied to cooperative games. As a proof-of-concept, we show preliminary results of reinforcement learning agents exploiting this environment.
%@ 978-3-030-30241-2
@inproceedings{cruz2019diplomacy,
abstract = {Reinforcement learning has been successfully applied to adversarial games, exhibiting its potential. However, most real-life scenarios also involve cooperation, in addition to competition. Using reinforcement learning in multi-agent cooperative games is, however, still mostly unexplored. In this paper, a reinforcement learning environment for the Diplomacy board game is presented, using the standard interface adopted by OpenAI Gym environments. Our main purpose is to enable straightforward comparison and reuse of existing reinforcement learning implementations when applied to cooperative games. As a proof-of-concept, we show preliminary results of reinforcement learning agents exploiting this environment.},
added-at = {2020-01-24T09:00:00.000+0100},
address = {Cham},
author = {Cruz, Diogo and Cruz, Jos{\'e} Aleixo and Lopes Cardoso, Henrique},
biburl = {https://www.bibsonomy.org/bibtex/2edf87af86f90fb18f7a175af7222475c/lanteunis},
booktitle = {Progress in Artificial Intelligence},
description = {Reinforcement Learning in Multi-agent Games: Open AI Gym Diplomacy Environment | SpringerLink},
editor = {Moura Oliveira, Paulo and Novais, Paulo and Reis, Lu{\'i}s Paulo},
interhash = {3bb79536f89d26f1a4858d64d14d3a66},
intrahash = {edf87af86f90fb18f7a175af7222475c},
isbn = {978-3-030-30241-2},
keywords = {ACKTR PPO environment multi_agent reinforcement_learning},
pages = {49--60},
publisher = {Springer International Publishing},
timestamp = {2020-01-25T13:15:15.000+0100},
title = {Reinforcement Learning in Multi-agent Games: Open AI Gym Diplomacy Environment},
year = 2019
}