A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
David Silver, Thomas Hubert, Julian Schrittwieser et al. · Science · 2018
The game of chess is the longest-studied domain in the history of artificial intelligence. The strongest programs are based on a combination of sophisticated search techniques, domain-specific adaptations, and handcrafted evaluation functions that have been refined by human experts over several decades. By contrast, the AlphaGo Zero program recently achieved superhuman performance in the game of Go by reinforcement learning from self-play. In this paper, we generalize this approach into a single AlphaZero algorithm that can achieve superhuman performance in many challenging games. Starting from random play and given no domain knowledge except the game rules, AlphaZero convincingly defeated a world champion program in the games of chess and shogi (Japanese chess), as well as Go.
Cited by 3,551 works. doi:10.1126/science.aar6404 Open-access PDF
Authors
- David Silver
- Thomas Hubert
- Julian Schrittwieser
- Ioannis Antonoglou
- Matthew Lai
- Arthur Guez
- Marc Lanctot
- Laurent Sifre
- Dharshan Kumaran
- Thore Graepel
- Timothy Lillicrap
- Karen Simonyan
- Demis Hassabis
References
- Mastering the game of Go with deep neural networks and tree search
- Identity Mappings in Deep Residual Networks
- Mastering the game of Go without human knowledge
- Some Studies in Machine Learning Using the Game of Checkers. II—Recent Progress
- An analysis of alpha-beta pruning
- Encyclopedia of Artificial Intelligence
- Is Learning The n-th Thing Any Easier Than Learning The First?
- TD-Gammon, a Self-Teaching Backgammon Program, Achieves Master-Level Play
- A Comparison of Prediction Accuracy, Complexity, and Training Time of Thirty-Three Old and New Classification Algorithms
- XXII. Programming a computer for playing chess