IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2010-03-11 11:05
Learning of Go board state evaluation by online adaptive natural gradient method
Hiroki Tomizawa, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2009-164
Abstract (in Japanese) (See Japanese page) 
(in English) We propose a supervised learning of Go board state evaluation function with many game records of experts'. By training a hierarchical neural network of which input is a board state of game records and output is binary (Black's win or White's win), the network is expected to output the expected winning probability (evaluation) of input board state.
Because it is difficult to learn large dimensional data such as Go board without overtraining, we restrict the dimension of parameters by taking into account of Go specific properties. Furthermore, we use online adaptive natural gradient method of which fisher information matrix is approximated by a block diagonal matrix to learn the huge number of data faster.
Keyword (in Japanese) (See Japanese page) 
(in English) Go / neural network / adaptive natural gradient method / / / / /  
Reference Info. IEICE Tech. Rep., vol. 109, no. 461, NC2009-164, pp. 449-454, March 2010.
Paper # NC2009-164 
Date of Issue 2010-03-02 (NC) 
ISSN Print edition: ISSN 0913-5685    Online edition: ISSN 2432-6380
Copyright
and
reproduction
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034)
Download PDF NC2009-164

Conference Information
Committee NC MBE  
Conference Date 2010-03-09 - 2010-03-11 
Place (in Japanese) (See Japanese page) 
Place (in English) Tamagawa University 
Topics (in Japanese) (See Japanese page) 
Topics (in English) General 
Paper Information
Registration To NC 
Conference Code 2010-03-NC-MBE 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Learning of Go board state evaluation by online adaptive natural gradient method 
Sub Title (in English)  
Keyword(1) Go  
Keyword(2) neural network  
Keyword(3) adaptive natural gradient method  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Hiroki Tomizawa  
1st Author's Affiliation Kyoto university (Kyoto Univ.)
2nd Author's Name Shin-ichi Maeda  
2nd Author's Affiliation Kyoto university (Kyoto Univ.)
3rd Author's Name Shin Ishii  
3rd Author's Affiliation Kyoto university (Kyoto Univ.)
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2010-03-11 11:05:00 
Presentation Time 25 minutes 
Registration for NC 
Paper # NC2009-164 
Volume (vol) vol.109 
Number (no) no.461 
Page pp.449-454 
#Pages
Date of Issue 2010-03-02 (NC) 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan