Hyper-parameter Optimization for Monte Carlo Tree Search using Self-play
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The Monte Carlo tree search (MCTS) is a popular method for implementing an intelligent game program. It has several hyper-parameters that require an optimization for showing the best performance. Due to the stochastic nature of the MCTS, the hyper-parameter optimization is difficult to solve. This paper uses the self-playing capability of the MCTS-based game program for optimizing the hyper-parameters. It seeks a winner path over the hyper-parameter space while performing the self-play. The top-q longest winners in the winner path compete for the final winner. The experiment using the 15-15-5 game (Omok in Korean name) showed a promising result.
Abstract
The Monte Carlo tree search (MCTS) is a popular method for implementing an intelligent game program. It has several hyper-parameters that require an optimization for showing the best performance. Due to the stochastic nature of the MCTS, the hyper-parameter optimization is difficult to solve. This paper uses the self-playing capability of the MCTS-based game program for optimizing the hyper-parameters. It seeks a winner path over the hyper-parameter space while performing the self-play. The top-q longest winners in the winner path compete for the final winner. The experiment using the 15-15-5 game (Omok in Korean name) showed a promising result.
참고문헌 (17)
- 최규혁, "FPS게임 플레이경험에 따른 플레이어의 시선경로 패턴 분석", 스마트미디어저널, 5(2), pp.33~41 (2016), (사)한국스마트미디어학회
- Walter, J. C., "An introduction to Monte Carlo methods", Physica A : Statistical Mechanics and its Applications, 418, pp.78~87 (2015)
- Browne, c., "A survey of Monte Carlo tree search methods", IEEE Transactions on Computational Intelligence and AI in Games, 4(1), pp.1~43 (2012)
- Fu, M. C., "Monte Carlo tree search: a tutorial", Proceedings of 2018Winter Simulation Conference, pp.222~236 (2018)
- Monte Carlo Go, Technical report, "Brügmann, B.", Department of Physics, Syracuse University, ~ (1993)
- Coulom, R., "Efficient selectivity and backup operators in Monte-Carlo tree search", 5th International Conference on Computers and Games, pp.72~83 (2006)
- Silver, D., "Mastering the game of Go with deep neural networks and tree search", Nature, 529, pp.484~489 (2016)
- Silver, D., "Mastering the game of Go without human knowledge", Nature, 550, pp.354~359 (2017)
- Poole, D. L., "Artificial Intelligence: Foundations of Computational Agents" (2017), Cambridge University Press
- Goodfellow, I., "Deep Learning" (2016), The MIT Press
- Bergstra, J. S., "Algorithms for hyper-parameter optimization", Advances in neural information processing systems, pp.2546~2554 (2011)
- Wang, H, "Analysis of hyper-parameter for small games:iterations or epochs in self-play?", arXiv:2003.05988v1, ~ (2020)
- Ruijl, B., "Combining simulated annealing and Monte Carlo Tree Search for expression simplification", Proceedings of the 6th International Conference on Agents and Artificial Intelligence, pp.724~731 (2014)
- 조우진, "네트워크 보안 관제를 위한 로그 시각화 방법", 스마트미디어저널, 7(4), pp.70~78 (2018), (사)한국스마트미디어학회[DOI]
- 서다솜, "딥러닝 설명을 위한 슈퍼픽셀 제외-포함 다중스케일 접근법", 스마트미디어저널, 8(2), pp.39~45 (2019), (사)한국스마트미디어학회[DOI]
- Li, L., "Hyperband: a novel bandit-based approach to hyperparameter optimization", The Journal of Machine Learning Research, 18(1), pp.1~52 (2017)
- Rakotoarison, H., "Automated machine learning with Monte Carlo tree search", IJCAI-19 28th International Joint Conference on Artificial Intelligence, pp.3296~3303 (2019)