# Accelerating Self-Play Learning in Go

arXiv: https://arxiv.org/abs/1902.10565 · HTML: https://mlsanity.com/paper/1902.10565/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Feb 2019.

## Built on by (most cited first, 30 of 64)

- [Scaling Scaling Laws with Board Games](https://mlsanity.com/paper/2104.03113/) (2021)
- [Modeling Strong and Human-Like Gameplay with KL-Regularized Search](https://mlsanity.com/paper/2112.07544/) (2021)
- [VideoWorld: Exploring Knowledge Learning from Unlabeled Videos](https://mlsanity.com/paper/2501.09781/) (2025)
- [Adversarial Policies Beat Superhuman Go AIs](https://mlsanity.com/paper/2211.00241/) (2022)
- [Superhuman Artificial Intelligence Can Improve Human Decision Making by Increasing Novelty](https://mlsanity.com/paper/2303.07462/) (2023)
- [LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios](https://mlsanity.com/paper/2310.08348/) (2023)
- [Mastering Chess with a Transformer Model](https://mlsanity.com/paper/2409.12272/) (2024)
- [Adaptive Informative Path Planning Using Deep Reinforcement Learning for UAV-based Active Sensing](https://mlsanity.com/paper/2109.13570/) (2021)
- [Are AlphaZero-like Agents Robust to Adversarial Perturbations?](https://mlsanity.com/paper/2211.03769/) (2022)
- [MiniZero: Comparative Analysis of AlphaZero and MuZero on Go, Othello, and Atari Games](https://mlsanity.com/paper/2310.11305/) (2023)
- [Monte Carlo Tree Search with Boltzmann Exploration](https://mlsanity.com/paper/2404.07732/) (2024)
- [Representation Matters for Mastering Chess: Improved Feature Representation in AlphaZero Outperforms Switching to Transformers](https://mlsanity.com/paper/2304.14918/) (2023)
- [Scalable Decision Making for Games of Imperfect Information](https://mlsanity.com/paper/2511.07312/) (2025)
- [Automatic Textbook Formalization](https://mlsanity.com/paper/2604.03071/) (2026)
- [Deep Learning for General Game Playing with Ludii and Polygames](https://mlsanity.com/paper/2101.09562/) (2021)
- [Transfer of Fully Convolutional Policy-Value Networks Between Games and Game Variants](https://mlsanity.com/paper/2102.12375/) (2021)
- [Adaptive Warm-Start MCTS in AlphaZero-like Deep Reinforcement Learning](https://mlsanity.com/paper/2105.06136/) (2021)
- [Explaining How a Neural Network Play the Go Game and Let People Learn](https://mlsanity.com/paper/2310.09838/) (2023)
- [Batch Monte Carlo Tree Search](https://mlsanity.com/paper/2104.04278/) (2021)
- [Learning to Find Proofs and Theorems by Learning to Refine Search Strategies: The Case of Loop Invariant Synthesis](https://mlsanity.com/paper/2205.14229/) (2022)
- [Deep Synoptic Monte Carlo Planning in Reconnaissance Blind Chess](https://mlsanity.com/paper/2110.01810/) (2021)
- [Increasing biases can be more efficient than increasing weights](https://mlsanity.com/paper/2301.00924/) (2023)
- [Multi-Agent Diagnostics for Robustness via Illuminated Diversity](https://mlsanity.com/paper/2401.13460/) (2024)
- [Improving Model and Search for Computer Go](https://mlsanity.com/paper/2102.03467/) (2021)
- [Spatial State-Action Features for General Games](https://mlsanity.com/paper/2201.06401/) (2022)
- [Impartial Games: A Challenge for Reinforcement Learning](https://mlsanity.com/paper/2205.12787/) (2022)
- [Targeted Search Control in AlphaZero for Effective Policy Improvement](https://mlsanity.com/paper/2302.12359/) (2023)
- [Score vs. Winrate in Score-Based Games: which Reward for Reinforcement Learning?](https://mlsanity.com/paper/2201.13176/) (2022)
- [BRExIt: On Opponent Modelling in Expert Iteration](https://mlsanity.com/paper/2206.00113/) (2022)
- [Vision Transformers for Computer Go](https://mlsanity.com/paper/2309.12675/) (2023)

## 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.