Suphx: Mastering Mahjong with Deep Reinforcement Learning

TL;DR

This paper introduces Suphx, a deep reinforcement learning-based AI that masters Mahjong, outperforming most top human players and achieving a significant milestone in complex multi-player imperfect-information game AI.

Contribution

Suphx is the first AI to outperform most top human Mahjong players, utilizing novel techniques like global reward prediction and run-time policy adaptation.

Findings

Suphx outperforms most top human players in Mahjong.

Suphx is rated above 99.99% of ranked players on Tenhou.

First AI to surpass top human Mahjong players.

Abstract

Artificial Intelligence (AI) has achieved great success in many domains, and game AI is widely regarded as its beachhead since the dawn of AI. In recent years, studies on game AI have gradually evolved from relatively simple environments (e.g., perfect-information games such as Go, chess, shogi or two-player imperfect-information games such as heads-up Texas hold'em) to more complex ones (e.g., multi-player imperfect-information games such as multi-player Texas hold'em and StartCraft II). Mahjong is a popular multi-player imperfect-information game worldwide but very challenging for AI research due to its complex playing/scoring rules and rich hidden information. We design an AI for Mahjong, named Suphx, based on deep reinforcement learning with some newly introduced techniques including global reward prediction, oracle guiding, and run-time policy adaptation. Suphx has demonstrated stronger performance than most top human players in terms of stable rank and is rated above 99.99% of all the officially ranked human players in the Tenhou platform. This is the first time that a computer program outperforms most top human players in Mahjong.