Abstract
Inspired by recent successes of Monte-Carlo tree search (MCTS) in a number of artificial intelligence (AI) application domains, we propose a reinforcement learning (RL) technique that iteratively applies MCTS on batches of small, finitehorizon versions of the original infinite-horizon Markov decision process. The terminal condition of the finite-horizon problems, or the leaf-node evaluator of the decision tree generated by MCTS, is specified using a combination of an estimated value function and an estimated policy function. The recommendations generated by the MCTS procedure are then provided as feedback in order to refine, through classification and regression, the leaf-node evaluator for the next iteration. We provide the first sample complexity bounds for a tree search-based RL algorithm. In addition, we show that a deep neural network implementation of the technique can create a competitive AI agent for the popular multi-player online battle arena (MOBA) game King of Glory.
Original language | English (US) |
---|---|
Title of host publication | 35th International Conference on Machine Learning, ICML 2018 |
Editors | Jennifer Dy, Andreas Krause |
Publisher | International Machine Learning Society (IMLS) |
Pages | 3572-3590 |
Number of pages | 19 |
ISBN (Electronic) | 9781510867963 |
State | Published - 2018 |
Event | 35th International Conference on Machine Learning, ICML 2018 - Stockholm, Sweden Duration: Jul 10 2018 → Jul 15 2018 |
Publication series
Name | 35th International Conference on Machine Learning, ICML 2018 |
---|---|
Volume | 5 |
Other
Other | 35th International Conference on Machine Learning, ICML 2018 |
---|---|
Country/Territory | Sweden |
City | Stockholm |
Period | 7/10/18 → 7/15/18 |
Funding
We wish to thank four anonymous reviewers, whose feedback helped to significantly improve the paper. We also thank our colleagues at Tencent AI Lab, particularly Carson Eisenach and Xiangru Lian, for technical help. Daniel Jiang is grateful for the support from Tencent AI Lab through a faculty award. The research of Han Liu was supported by NSF CAREER Award DMS1454377, NSF IIS1408910, and NSF IIS1332109. This material is also based upon work supported by the National Science Foundation under grant no. 1740762 "Collaborative Research: TRIPODS Institute for Optimization and Learning."
ASJC Scopus subject areas
- Computational Theory and Mathematics
- Human-Computer Interaction
- Software