Presentation 2007-06-15
Solving Sudoku Puzzle with Neural Network
Akihiro OHTANI, Satoshi MATSUDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Based on the insight that all the intellectual activities of human beings are carried out as optimization process, we already proposed Hopfield network model of decision making. Furthermore, we have an interest in what kind of inferences Hopfield network can do and cannot in problem solving, and that there is the similarities in problem solving abilities between humans and Hopfield networks. Solving puzzles is also a typical intellectual activity, so, in this paper, by trying to solve Sudoku puzzles with optimizing Hopfield networks, we show that Hopfield network can do "scanning" technique, but not "forcing chains" technique. This is an interesting phenomenon because humans feel Sudoku problems, which only require scanning technique, easy, but ones, which require forcing chains techniques, hard. We thus show that there is the similarities in problem solving abilities between humans and Hopfield network.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Neural Network / Hopfield Network / Combinational Optimization Problem
Paper # NC2007-20
Date of Issue

Conference Information
Committee NC
Conference Date 2007/6/7(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Solving Sudoku Puzzle with Neural Network
Sub Title (in English)
Keyword(1) Neural Network
Keyword(2) Hopfield Network
Keyword(3) Combinational Optimization Problem
1st Author's Name Akihiro OHTANI
1st Author's Affiliation Department of Mathematical Information Engineering, College of Industrial Technology, Nihon University()
2nd Author's Name Satoshi MATSUDA
2nd Author's Affiliation Department of Mathematical Information Engineering, College of Industrial Technology, Nihon University
Date 2007-06-15
Paper # NC2007-20
Volume (vol) vol.107
Number (no) 92
Page pp.pp.-
#Pages 6
Date of Issue