ELF OpenGo: An Analysis and Open Reimplementation of AlphaZero

TL;DR

ELF OpenGo is an open-source implementation of AlphaZero that achieves superhuman Go performance, enabling extensive analysis and facilitating further research in deep reinforcement learning.

Contribution

It provides the first open-source Go AI with superhuman performance and offers detailed analysis through ablation studies and gameplay insights.

Findings

Achieves 20:0 record against top professionals

Enables extensive ablation and analysis of training and inference

Provides publicly available code, models, and datasets

Abstract

The AlphaGo, AlphaGo Zero, and AlphaZero series of algorithms are remarkable demonstrations of deep reinforcement learning's capabilities, achieving superhuman performance in the complex game of Go with progressively increasing autonomy. However, many obstacles remain in the understanding of and usability of these promising approaches by the research community. Toward elucidating unresolved mysteries and facilitating future research, we propose ELF OpenGo, an open-source reimplementation of the AlphaZero algorithm. ELF OpenGo is the first open-source Go AI to convincingly demonstrate superhuman performance with a perfect (20:0) record against global top professionals. We apply ELF OpenGo to conduct extensive ablation studies, and to identify and analyze numerous interesting phenomena in both the model training and in the gameplay inference procedures. Our code, models, selfplay datasets, and auxiliary data are publicly available at https://ai.facebook.com/tools/elf-opengo/.