Edinburgh Research Explorer

Training Deep Convolutional Neural Networks to Play Go

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Related Edinburgh Organisations

Open Access permissions

Open

Original languageEnglish
Title of host publicationProceedings of the 32nd International Conference on Machine Learning (IMCL 2015)
Place of PublicationLille, France
PublisherPMLR
Pages1766-1774
Number of pages9
Volume37
Publication statusPublished - 2015
Event32nd international conference on machine learning - Lille, France
Duration: 6 Jul 201511 Jul 2015
https://icml.cc/2015/

Conference

Conference32nd international conference on machine learning
Abbreviated titleICML 2015
CountryFrance
CityLille
Period6/07/1511/07/15
Internet address

Abstract

Mastering the game of Go has remained a longstanding challenge to the field of AI. Modern computer Go systems rely on processing millions of possible future positions to play well,but intuitively a stronger and more ‘human like’ way to play the game would be to rely on pattern recognition abilities rather then brute force computation. Following this sentiment, we train deep convolutional neural networks to play Go by training them to predict the moves made by expert Go players. To solve this problem we introduce a number of novel techniques, including a method of tying weights in the network to ‘hard code’ symmetries that are expect to exist in the target function, and demonstrate in anablation study they considerably improve performance.Our final networks are able to achieve move prediction accuracies of 41.1% and 44.4%on two different Go datasets, surpassing previous state of the art on this task by significant margins.Additionally, while previous move prediction programs have not yielded strong Go playing programs, we show that the networks trained inthis work acquired high levels of skill. Our convolutional neural networks can consistently defeat the well known Go program GNU Go, indicating it is state of the art among programs that do not use Monte Carlo Tree Search. It is also able to win some games against state of the artGo playing program Fuego while using a fraction of the play time. This success at playing Go indicates high level principles of the game we relearned.

Event

32nd international conference on machine learning

6/07/1511/07/15

Lille, France

Event: Conference

Download statistics

No data available

ID: 24449733