About this document
Suphx: AI Mastery in Mahjong by Kitanovic Nenad is a document available to read on EtoBox.
This document describes an AI system called Suphx for playing the multiplayer imperfect-information game of Mahjong. Suphx uses deep reinforcement learning with techniques like global reward prediction, oracle guiding, and run-time policy adaptation. Suphx demonstrates superhuman performance, outperforming most top human players by achieving a stable rank above 99.99% of officially ranked human players on the popular Tenhou platform. This is the first time a computer program has outperformed most top human
- Author
- Kitanovic Nenad
- Language
- EN