# Accelerating Self-Play Learning in Go

DevFeed: [Accelerating Self-Play Learning in Go](<https://devfeed.tech/articles/accelerating-self-play-learning-in-go-20147.md>)

Original publisher: [Read original article](<https://blog.janestreet.com/accelerating-self-play-learning-in-go/>)

Author: David Wu

Published: 2019-02-28T00:00:00Z

Content type: article

Language: en

Sources: [Jane Street](<https://devfeed.tech/sources/jane-street.md>)

Topics: [Go Language](<https://devfeed.tech/topics/go-language.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Deep learning](<https://devfeed.tech/topics/deep-learning.md>), [Deep neural networks](<https://devfeed.tech/topics/deep-neural-networks.md>), [AI Research](<https://devfeed.tech/topics/ai-research.md>)

Tags: [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [go](<https://devfeed.tech/tags/go.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [neural](<https://devfeed.tech/tags/neural.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [research](<https://devfeed.tech/tags/research.md>)

## AI overview

Jane Street describes a personal research project applying neural-network training and self-play learning to Go. Short and medium-length runs produced strong professional or possibly superhuman play, and the team released a paper, source code, trained networks, and an online bot. The article reports preliminary estimates that the techniques may accelerate learning compared with Leela Zero, while noting that further testing is needed.

## Source excerpt

At Jane Street, over the last few years, we've been increasingly exploring machine learning to improve our models. Many of us are fascinated by the rapid improvement we see in a wide variety of applications due to developments in deep learning and reinforcement learning, both for its exciting potential for our own problems, and also on a personal level of pure interest and curiosity outside of work.