Presentation 1994/7/8
Learning of Complex Pattern Concepts on a Game Board by the Analysis of the Cause of Failure
Morihiko Tajima, Noriaki Sanechika,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes the new result produced by our learning system named RLS(Recursive Learning System).RLS-Othello is the game playing and learning system on RLs.It learns useful concepts by analyzing the cause of failure from a record of a game.RLS also utilizes EBG as an auxiliary means.Because of the shortage of the knowledge for EBG,RLS-Othello has not been able to learn very useful concepts including complex patterns.In this paper,we introduced variables denoting time into goal concept,domain theory, and training example,and enabled RLS-Othello to describe the relations between any two moments.By the improvement of RLS- Othello,RLS-Othello became to be able to learn the concept of "pai r wing" which is one of the complex patterns of the game.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) game / game learning / Othello game / EBG / RLS / domain theory
Paper # AI94-29
Date of Issue

Conference Information
Committee AI
Conference Date 1994/7/8(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Learning of Complex Pattern Concepts on a Game Board by the Analysis of the Cause of Failure
Sub Title (in English)
Keyword(1) game
Keyword(2) game learning
Keyword(3) Othello game
Keyword(4) EBG
Keyword(5) RLS
Keyword(6) domain theory
1st Author's Name Morihiko Tajima
1st Author's Affiliation Electrotechnical Laboratory()
2nd Author's Name Noriaki Sanechika
2nd Author's Affiliation Electrotechnical Laboratory
Date 1994/7/8
Paper # AI94-29
Volume (vol) vol.94
Number (no) 133
Page pp.pp.-
#Pages 8
Date of Issue