Reinforcement learning of intelligent characters in fighting action games

Authors:
Byeong Heon Cho;Sung Hoon Jung;Kwang-Hyun Shim;Yeong Rak Seong;Ha Ryoung Oh
Affiliations:
Digital Content Research Division, ETRI, Daejeon, Korea;Dept. of Information and Comm. Eng., Hansung Univ., Seoul, Korea;Digital Content Research Division, ETRI, Daejeon, Korea;School of Electrical Engineering, Kookmin Univ., Seoul, Korea;School of Electrical Engineering, Kookmin Univ., Seoul, Korea
Venue:
ICEC'06 Proceedings of the 5th international conference on Entertainment Computing
Year:
2006

Citing 2
Cited 1

Exploiting Intelligence in Fighting Action Games Using Neural Networks

IEICE - Transactions on Information and Systems
Adaptation of intelligent characters to changes of game environments

CIS'05 Proceedings of the 2005 international conference on Computational Intelligence and Security - Volume Part I

Comparison of AI techniques for fighting action games - genetic algorithms/neural networks/evolutionary neural networks

ICEC'07 Proceedings of the 6th international conference on Entertainment Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we investigate reinforcement learning (RL) of intelligent characters, based on neural network technology, for fighting action games. RL can be either on-policy or off-policy. We apply both schemes to tabula rasa learning and adaptation. The experimental results show that (1) in tabula rasa leaning, off-policy RL outperforms on-policy RL, but (2) in adaptation, on-policy RL outperforms off-policy RL.