Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

TLDR

Chess AI has relied on handcrafted search and evaluation, while AlphaGo Zero demonstrated that tabula rasa reinforcement learning can reach superhuman performance in Go. This work proposes AlphaZero, a general algorithm that learns from self‑play without domain knowledge to achieve superhuman play across multiple games. AlphaZero extends AlphaGo Zero by applying deep reinforcement learning with self‑play to any game defined by its rules, using a unified neural network for policy and value. Starting from random play, AlphaZero reached superhuman performance in chess, shogi, and Go within 24 hours and defeated a world‑champion program in each game.

Abstract

The game of chess is the most widely-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. In contrast, the AlphaGo Zero program recently achieved superhuman performance in the game of Go, by tabula rasa reinforcement learning from games of self-play. In this paper, we generalise this approach into a single AlphaZero algorithm that can achieve, tabula rasa, superhuman performance in many challenging domains. Starting from random play, and given no domain knowledge except the game rules, AlphaZero achieved within 24 hours a superhuman level of play in the games of chess and shogi (Japanese chess) as well as Go, and convincingly defeated a world-champion program in each case.