mlsanityData to 7 Oct 2026

Accelerating Self-Play Learning in Go

arXiv 1902.10565 ↗First posted 27 Feb 2019cs.LG stat.MLcited by 60

Not on the map. Why?

Abstract (arXiv)

By introducing several improvements to the AlphaZero process and architecture, we greatly accelerate self-play learning in Go, achieving a 50x reduction in computation over comparable methods. Like AlphaZero and replications such as ELF OpenGo and Leela Zero, our bot KataGo only learns from neural-net-guided Monte Carlo tree search self-play. But whereas AlphaZero required thousands of TPUs over several days and ELF required thousands of GPUs over two weeks, KataGo surpasses ELF's final model after only 19 days on fewer than 30 GPUs. Much of the speedup involves non-domain-specific improvements that might directly transfer to other problems. Further gains from domain-specific techniques reveal the remaining efficiency gap between the best methods and purely general methods such as AlphaZero. Our work is a step towards making learning in state spaces as large as Go possible without large-scale computational resources.

Key facts. Cited by 60 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 10; 2022: 11; 2023: 13; 2024: 8; 2025: 8; 2026: 10.

60 papers.

Sort:

By area:

  1. 2021Scaling Scaling Laws with Board Gamescited by 62
  2. 2021Modeling Strong and Human-Like Gameplay with KL-Regularized Searchcited by 50
  3. 2025VideoWorld: Exploring Knowledge Learning from Unlabeled Videoscited by 42
  4. 2022Adversarial Policies Beat Superhuman Go AIscited by 29
  5. 2023Superhuman Artificial Intelligence Can Improve Human Decision Making by Increasing Noveltycited by 27
  6. 2023LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarioscited by 20
  7. 2022Are AlphaZero-like Agents Robust to Adversarial Perturbations?cited by 14
  8. 2023MiniZero: Comparative Analysis of AlphaZero and MuZero on Go, Othello, and Atari Gamescited by 14
  9. 2024Mastering Chess with a Transformer Modelcited by 14
  10. 2021Adaptive Informative Path Planning Using Deep Reinforcement Learning for UAV-based Active Sensingcited by 12
  11. 2024Monte Carlo Tree Search with Boltzmann Explorationcited by 12
  12. 2023Representation Matters for Mastering Chess: Improved Feature Representation in AlphaZero Outperforms Switching to Transformerscited by 8
38 more
  1. 2021Deep Learning for General Game Playing with Ludii and Polygamescited by 7
  2. 2025Scalable Decision Making for Games of Imperfect Informationcited by 7
  3. 2021Transfer of Fully Convolutional Policy-Value Networks Between Games and Game Variantscited by 6
  4. 2021Adaptive Warm-Start MCTS in AlphaZero-like Deep Reinforcement Learningcited by 6
  5. 2023Explaining How a Neural Network Play the Go Game and Let People Learncited by 6
  6. 2021Batch Monte Carlo Tree Searchcited by 5
  7. 2022Learning to Find Proofs and Theorems by Learning to Refine Search Strategies: The Case of Loop Invariant Synthesiscited by 5
  8. 2026Automatic Textbook Formalizationcited by 5
  9. 2021Deep Synoptic Monte Carlo Planning in Reconnaissance Blind Chesscited by 4
  10. 2023Increasing biases can be more efficient than increasing weightscited by 4
  11. 2024Multi-Agent Diagnostics for Robustness via Illuminated Diversitycited by 4
  12. 2021Improving Model and Search for Computer Gocited by 3
  13. 2022Spatial State-Action Features for General Gamescited by 3
  14. 2022Impartial Games: A Challenge for Reinforcement Learningcited by 3
  15. 2023Targeted Search Control in AlphaZero for Effective Policy Improvementcited by 3
  16. 2022Score vs. Winrate in Score-Based Games: which Reward for Reinforcement Learning?cited by 2
  17. 2022BRExIt: On Opponent Modelling in Expert Iterationcited by 2
  18. 2023Vision Transformers for Computer Gocited by 2
  19. 2023ADAPTER-RL: Adaptation of Any Agent using Reinforcement Learningcited by 2
  20. 2024ReZero: Boosting MCTS-based Algorithms by Backward-view and Entire-buffer Reanalyzecited by 2
  21. 2022PGD: A Large-scale Professional Go Dataset for Data-driven Analyticscited by 1
  22. 2023The Mathematical Gamecited by 1
  23. 2023Towards Attributions of Input Variables in a Coalitioncited by 1
  24. 2025Empowering LLMs in Decision Games through Algorithmic Data Synthesiscited by 1
  25. 2025Robust Agents in Open-Ended Worldscited by 1
  26. 2026Mixing Expert Knowledge: Bring Human Thoughts Back To the Game of Gocited by 1
  27. 2026Decodable but Not Faithful: Coupling Natural-Language Rationales to Programmatic Verifierscited by 1
  28. 2026Matilda: Engine-Agnostic Search with Human Policy Guidancecited by 1
  29. 2021OLIVAW: Mastering Othello without Human Knowledge, nor a Fortunecited by 0
  30. 2022The cost of passing -- using deep learning AIs to expand our understanding of the ancient game of Gocited by 0
  31. 2022Graph Value Iterationcited by 0
  32. 2022The ProfessionAl Go annotation datasEt (PAGE)cited by 0
  33. 2023Know your Enemy: Investigating Monte-Carlo Tree Search with Opponent Models in Pommermancited by 0
  34. 2023Agents Explore the Environment Beyond Good Actions to Improve Their Model for Better Decisionscited by 0
  35. 2024Monte Carlo Search Algorithms Discovering Monte Carlo Tree Search Exploration Termscited by 0
  36. 2024AlphaViT: A flexible game-playing AI for multiple games and variable board sizescited by 0
  37. 2024Mastering Chinese Chess AI (Xiangqi) Without Searchcited by 0
  38. 2024Bridging Local and Global Knowledge via Transformer in Board Gamescited by 0

50 most cited of 60 All 60 (JSON)