AlphaZero: Shedding new light on chess, shogi, and Go
In late 2017 we introduced AlphaZero, a single system that taught itself from scratch how to master the games of chess, shogi (Japanese chess), and Go, beating a world-champion program in each case. We were excited by the preliminary results and thrilled to see the response from members of the chess community, who saw in AlphaZero’s games a ground-breaking, highly dynamic and “unconventional” style of play that differed from any chess playing engine that came before it.
December 6, 2018 Research David Silver, Thomas Hubert, Julian Schrittwieser, Demis Hassabis In late 2017 we introduced AlphaZero, a single system that taught itself from scratch how to master the games of chess, shogi(Japanese chess), and Go, beating a world-champion program in each case. We were excited by the preliminary results and thrilled to see the response from members of the chess community, who saw in AlphaZero’s games a ground-breaking, highly dynamic and “unconventional” style of play that differed from any chess playing engine that came before it. Today, we are delighted to…
related reading
- A Very Unlikely Chess Game | Slate Star Codexslatestarcodex.com
- deepchess.pdfcs.tau.ac.il
- Grandmaster-Level Chess Without Searcharxiv.org
- Simple Alpha Zeroweb.stanford.edu
- After AI beat them, professional go players got better and more creativehenrikkarlsson.xyz
- How Go Players Disempower Themselves to AI — LessWronglesswrong.com
- Playing chess with large language modelsnicholas.carlini.com
- Chess Engines: A Zero to One Guidechessengines.super.site
- AlphaGo versus Lee Sedol - Wikipediaen.wikipedia.org
- AlphaGo Zero: Minimal Policy Improvement, Expectation Propagation and other Connectionsinference.vc
- Eric Jang (@ericjang11) on Xx.com
- Generally capable agents emerge from open-ended playdeepmind.google