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

**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  
**arXiv:** [1712.01815](https://arxiv.org/abs/1712.01815)  
**Date:** 2017-12-05  
**Summary page:** https://www.summarizepaper.com/en/paper/1712.01815v1/

## Key points

- Paper titled "Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm"
- 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,and Demis Hassabis
- Discusses the success of artificial intelligence in chess through reinforcement learning algorithms like AlphaZero
- AlphaZero achieved superhuman performance in chess without prior knowledge through self-play
- Demonstrated excellence in other complex domains such as shogi and Go within 24 hours
- Showcases the power of tabula rasa approaches in AI capabilities without domain-specific expertise or pre-existing data

## Summary

The paper "Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm," authored by David Silver, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran,Thore Graepel,Timothy Lillicrap,Karen Simonyan,and Demis Hassabis,discusses the groundbreaking success of artificial intelligence in the domain of chess. Chess has been extensively studied in AI history and the best programs have relied on advanced search techniques and manually crafted evaluation functions. However,the authors highlight the transformative potential of reinforcement learning algorithms like AlphaZero which achieved superhuman performance in chess through self-play without any prior knowledge. This approach was further demonstrated to excel in other complex domains such as shogi (Japanese chess) and Go within 24 hours. The study showcases the power of tabula rasa approaches like AlphaZero in pushing the boundaries of AI capabilities and achieving remarkable proficiency without relying on domain-specific expertise or pre-existing data. , , , , .