Abalearn: a risk-sensitive approach to self-play learning in Abalone

Campos, Pedro; Langlois, Thibault

http://hdl.handle.net/10400.13/4603

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
Abalearn.pdf		210.75 KB	Adobe PDF	Download

Send Feedback

Authors

Campos, Pedro

Langlois, Thibault

Abstract(s)

This paper presents Abalearn, a self-teaching Abalone pro gram capable of automatically reaching an intermediate level of play without needing expert-labeled training examples, deep searches or ex posure to competent play. Our approach is based on a reinforcement learning algorithm that is risk seeking, since defensive players in Abalone tend to never end a game. We show that it is the risk-sensitivity that allows a successful self-play training. We also propose a set of features that seem relevant for achiev ing a good level of play. We evaluate our approach using a fixed heuristic opponent as a bench mark, pitting our agents against human players online and comparing samples of our agents at different times of training.

Keywords

Abalearn Self-play learning Abalone . Faculdade de Ciências Exatas e da Engenharia

URI

http://hdl.handle.net/10400.13/4603

Citation

Campos, P., Langlois, T. (2003). Abalearn: A Risk-Sensitive Approach to Self-play Learning in Abalone. In: Lavrač, N., Gamberger, D., Blockeel, H., Todorovski, L. (eds) Machine Learning: ECML 2003. ECML 2003. Lecture Notes in Computer Science(), vol 2837. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-39857-8_6