Deep Reinforcement Learning for General Video Game AI

被引：0

作者：

Tornado, Ruben Rodriguez ^{[1
]}

Bontrager, Philip ^{[1
]}

Togelius, Julian ^{[1
]}

Liu, Jialin ^{[2
]}

Perez-Liebana, Diego ^{[3
]}

机构：

[1] NYU, New York, NY 10012 USA

[2] Southern Univ Sci & Technol, Shenzhen, Peoples R China

[3] Queen Mary Univ London, London, England

来源：

PROCEEDINGS OF THE 2018 IEEE CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND GAMES (CIG'18) | 2018年

关键词：

deep reinforcement learning; general video game AI; video game description language; OpenAI Gym; advantage actor critic; deep Q-learning; NEURAL-NETWORKS; GO;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The General Video Game AI (GVGAI) competition and its associated software framework provides a way of benchmarking AI algorithms on a large number of games written in a domain-specific description language. While the competition has seen plenty of interest, it has so far focused on online planning, providing a forward model that allows the use of algorithms such as Monte Carlo Tree Search. In this paper, we describe how we interface GVGAI to the OpenAI Gym environment, a widely used way of connecting agents to reinforcement learning problems. Using this interface, we characterize how widely used implementations of several deep reinforcement learning algorithms fare on a number of GVGAI games. We further analyze the results to provide a first indication of the relative difficulty of these games relative to each other, and relative to those in the Arcade Learning Environment under similar conditions.

引用

页码：316 / 323

页数：8

共 50 条

[11] Accelerating deep reinforcement learning model for game strategy
Li, Yifan
Fang, Yuchun
Akhtar, Zahid
NEUROCOMPUTING, 2020, 408 : 157 - 168
[12] Grid-Wise Control for Multi-Agent Reinforcement Learning in Video Game AI
Han, Lei
Sun, Peng
Du, Yali
Xiong, Jiechao
Wang, Qing
Sun, Xinghai
Liu, Han
Zhang, Tong
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 97, 2019, 97
[13] Deep RTS: A Game Environment for Deep Reinforcement Learning in Real-Time Strategy Games
Andersen, Per-Arne
Goodwin, Morten
Granmo, Ole-Christoffer
PROCEEDINGS OF THE 2018 IEEE CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND GAMES (CIG'18), 2018, : 149 - 156
[14] Train timetabling with the general learning environment and multi-agent deep reinforcement learning
Li, Wenqing
Ni, Shaoquan
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL, 2022, 157 : 230 - 251
[15] Deep Learning for Video Game Playing
Justesen, Niels
Bontrager, Philip
Togelius, Julian
Risi, Sebastian
IEEE TRANSACTIONS ON GAMES, 2020, 12 (01) : 1 - 20
[16] From Reinforcement Learning to Deep Reinforcement Learning: An Overview
Agostinelli, Forest
Hocquet, Guillaume
Singh, Sameer
Baldi, Pierre
BRAVERMAN READINGS IN MACHINE LEARNING: KEY IDEAS FROM INCEPTION TO CURRENT STATE, 2018, 11100 : 298 - 328
[17] Comparison of Deep Reinforcement Learning Approaches for Intelligent Game Playing
Jeerige, Anoop
Bein, Doina
Verma, Abhishek
2019 IEEE 9TH ANNUAL COMPUTING AND COMMUNICATION WORKSHOP AND CONFERENCE (CCWC), 2019, : 366 - 371
[18] General game-playing and reinforcement learning
Levinson, R
COMPUTATIONAL INTELLIGENCE, 1996, 12 (01) : 155 - 176
[19] A Deep Reinforcement Learning Blind AI in DareFightingICE
Nguyen, Thai Van
Dai, Xincheng
Khan, Ibrahim
Thawonmas, Ruck
2022 IEEE CONFERENCE ON GAMES, COG, 2022, : 632 - 637
[20] Deep Reinforcement Learning for FlipIt Security Game
Greige, Laura
Chin, Peter
COMPLEX NETWORKS & THEIR APPLICATIONS X, VOL 1, 2022, 1015 : 831 - 843

← 1 2 3 4 5 →