Targeted Search Control in AlphaZero for Effective Policy Improvement
Alexandre Trudeau, Michael Bowling

TL;DR
This paper introduces Go-Exploit, a novel search control strategy for AlphaZero that enhances exploration and value learning by starting self-play from diverse states, leading to improved sample efficiency and stronger gameplay in board games.
Contribution
Go-Exploit's approach of sampling initial states from an archive and shorter trajectories improves exploration, value generalization, and sample efficiency in AlphaZero.
Findings
Go-Exploit outperforms standard AlphaZero in sample efficiency.
It achieves stronger performance against reference opponents.
It surpasses KataGo's sample efficiency and effectiveness.
Abstract
AlphaZero is a self-play reinforcement learning algorithm that achieves superhuman play in chess, shogi, and Go via policy iteration. To be an effective policy improvement operator, AlphaZero's search requires accurate value estimates for the states appearing in its search tree. AlphaZero trains upon self-play matches beginning from the initial state of a game and only samples actions over the first few moves, limiting its exploration of states deeper in the game tree. We introduce Go-Exploit, a novel search control strategy for AlphaZero. Go-Exploit samples the start state of its self-play trajectories from an archive of states of interest. Beginning self-play trajectories from varied starting states enables Go-Exploit to more effectively explore the game tree and to learn a value function that generalizes better. Producing shorter self-play trajectories allows Go-Exploit to train upon…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsArtificial Intelligence in Games · Sports Analytics and Performance · Educational Games and Gamification
MethodsAlphaZero
