Title
Solving Very Difficult Japanese Puzzles with a Hybrid Evolutionary-Logic Algorithm
Abstract
In this paper we present a hybrid evolutionary algorithm to solve a popular logic-type puzzle, the so called Japanese puzzle. We propose to use the evolutionary algorithm in order to initialize a logic ad-hoc algorithm, which works as a local search and implicitly defines the fitness function of the problem. Two novel operators, one for initializing the evolutionary algorithm and a second one providing a novel type of mutation adapted to Japanese puzzles are described in the paper.
Year
DOI
Venue
2008
10.1007/978-3-540-89694-4_37
SEAL
Keywords
Field
DocType
hybrid evolutionary-logic algorithm,difficult japanese puzzles,fitness function,novel type,hybrid evolutionary algorithm,popular logic-type puzzle,novel operator,japanese puzzle,logic ad-hoc algorithm,evolutionary algorithm,local search
Interactive evolutionary computation,Mathematical optimization,Evolutionary algorithm,Computer science,Algorithm,Fitness function,Operator (computer programming),Artificial intelligence,Local search (optimization),Cultural algorithm,Evolutionary programming,Evolutionary music
Conference
Volume
ISSN
Citations 
5361
0302-9743
1
PageRank 
References 
Authors
0.37
3
5
Name
Order
Citations
PageRank
Emilio G. Ortiz-García114010.54
Sancho Salcedo-Sanz258071.21
Ángel M. Pérez-Bellido313011.71
Antonio Portilla-Figueras414719.07
Xin Yao514858945.63