Grandmaster-Level Chess Without Search

N/A

Amortized Planning with Large-Scale Transformers: A Case Study on Chess Anian Ruoss∗1 Grégoire Delétang∗1 Sourabh Medapati1 Jordi Grau-Moya1 Li Kevin Wenliang1 Elliot Catt1 John Reid1 Cannada A. Lewis2 Joel Veness1 arXiv:2402.04494v2 [cs.LG] 21 Oct 2024 Tim Genewein1…

saved by

• Braeden Hall
• Vincent Huang

related reading

• Playing chess with large language modelsnicholas.carlini.com
• Learning Chess With Language Models and Transformersarxiv.org
• Learning to Play Chess from Textbooks (LEAP): a Corpus for Evaluating Chess Moves based on Sentiment Analysisarxiv.org
• Chess Engines: A Zero to One Guidechessengines.super.site
• deepchess.pdfcs.tau.ac.il
• LLM Chess: Benchmarking Reasoning and Instruction-Following in LLMs through Chessarxiv.org
• Something weird is happening with LLMs and chessdynomight.substack.com
• Chess-GPT’s Internal World Model | Adam Karvonenadamkarvonen.github.io
• Stream of Search (SoS): Learning to Search in Languagearxiv.org
• Composer2.pdfcursor.com
• A Very Unlikely Chess Game | Slate Star Codexslatestarcodex.com
• Transformer Progress | Leela Chess Zerolczero.org