The Arcade Learning Environment: An Evaluation Platform for General Agents

被引:1204
作者
Bellemare, Marc G. [1 ]
Naddaf, Yavar [2 ]
Veness, Joel [1 ]
Bowling, Michael [1 ]
机构
[1] Univ Alberta, Edmonton, AB, Canada
[2] Empir Results Inc, Vancouver, BC, Canada
关键词
Reinforcement learning;
D O I
10.1613/jair.3912
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this article we introduce the Arcade Learning Environment (ALE): both a challenge problem and a platform and methodology for evaluating the development of general, domain-independent AI technology. ALE provides an interface to hundreds of Atari 2600 game environments, each one different, interesting, and designed to be a challenge for human players. ALE presents significant research challenges for reinforcement learning, model learning, model-based planning, imitation learning, transfer learning, and intrinsic motivation. Most importantly, it provides a rigorous testbed for evaluating and comparing approaches to these problems. We illustrate the promise of ALE by developing and benchmarking domain-independent agents designed using well-established AI techniques for both reinforcement learning and planning. In doing so, we also propose an evaluation methodology made possible by ALE, reporting empirical results on over 55 different games. All of the software, including the benchmark agents, is publicly available.
引用
收藏
页码:253 / 279
页数:27
相关论文
共 32 条
[1]  
[Anonymous], 2008, THESIS U LUGANO
[2]  
[Anonymous], P GEN EV COMP C GECC
[3]  
Bellemare M., 2012, P 26 C ART INT AAAI
[4]   A Survey of Monte Carlo Tree Search Methods [J].
Browne, Cameron B. ;
Powley, Edward ;
Whitehouse, Daniel ;
Lucas, Simon M. ;
Cowling, Peter I. ;
Rohlfshagen, Philipp ;
Tavener, Stephen ;
Perez, Diego ;
Samothrakis, Spyridon ;
Colton, Simon .
IEEE TRANSACTIONS ON COMPUTATIONAL INTELLIGENCE AND AI IN GAMES, 2012, 4 (01) :1-43
[5]  
Cobo L. C, 2011, P 22 2 INT JOINT C A
[6]   A Survey of the Seventh International Planning Competition [J].
Coles, Amanda ;
Coles, Andrew ;
Garcia Olaya, Angel ;
Jimenez, Sergio ;
Linares Lopez, Carlos ;
Sanner, Scott ;
Yoon, Sungwook .
AI MAGAZINE, 2012, 33 (01) :83-88
[7]  
Dowe D. L., 1998, P INT C COMP INT MUL
[8]  
Genesereth M, 2005, AI MAG, V26, P62
[9]  
Gionis A, 1999, PROCEEDINGS OF THE TWENTY-FIFTH INTERNATIONAL CONFERENCE ON VERY LARGE DATA BASES, P518
[10]   Measuring universal intelligence: Towards an anytime intelligence test [J].
Hernandez-Orallo, Jose ;
Dowe, David L. .
ARTIFICIAL INTELLIGENCE, 2010, 174 (18) :1508-1539