Deep Reinforcement Learning for General Video Game AI

Torrado，Ruben Rodriguez1; Bontrager，Philip1; Togelius，Julian1; Liu，Jialin2; Perez-Liebana，Diego3

10.1109/CIG.2018.8490422

2018-10-11

2325-4270

2325-4289

978-1-5386-4360-0

IEEE Conference on Computatonal Intelligence and Games, CIG

2018-August

1-8

14-17 Aug. 2018

Maastricht, Netherlands

IEEE Computer Society

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.

Advantage actor critic Deep Q-learning Deep reinforcement learning General video game AI OpenAI Gym Video game description language

其他

英语

EI

Ministry of Science and Technology of the People's Republic of China[2017YFC0804003]

20184706125304

Artificial intelligence ; Computer aided instruction ; Computer programming ; Deep learning ; Human computer interaction ; Learning algorithms

Computer Software, Data Handling and Applications:723

2-s2.0-85056836621

Scopus

1.New York University, ,New York,United States
2.Southern University of Science and Technology, ,Shenzhen,China
3.Queen Mary University of London, ,London,United Kingdom

Torrado，Ruben Rodriguez,Bontrager，Philip,Togelius，Julian,et al. Deep Reinforcement Learning for General Video Game AI[C]:IEEE Computer Society,2018:1-8.