Title :
Learning to play Tic-tac-toe
Author :
Widyantoro, Dwi H. ; Vembrina, Yus G.
Author_Institution :
Sch. of Electr. Eng. & Inf., Inst. of Technol. Bandung, Bandung, Indonesia
Abstract :
This paper reports our experiment on applying Q Learning algorithm for learning to play Tic-tac-toe. The original algorithm is modified by updating the Q value only when the game terminates, propagating the update process from the final move backward to the first move, and incorporating a new update rule. We evaluate the agent performance using full-board and partial-board representations. In this evaluation, the agent plays the tic-tac-toe game against human players. The evaluation results show that the performance of modified Q Learning algorithm with partial-board representation is comparable to that of human players.
Keywords :
learning (artificial intelligence); Q learning algorithm; agent performance; full-board representations; partial-board representations; tic-tac-toe; Delay; Function approximation; Humans; Informatics; Machine learning; Machine learning algorithms; Mobile robots; Production facilities; Robot control; State-space methods; Board-Game; Q Learning; Tic-tac-toe;
Conference_Titel :
Electrical Engineering and Informatics, 2009. ICEEI '09. International Conference on
Conference_Location :
Selangor
Print_ISBN :
978-1-4244-4913-2
DOI :
10.1109/ICEEI.2009.5254776