Simulated annealing approach to solve nonogram puzzles with multiple solutions

Wen-li Wang, Mei Huei Tang

Research output: Contribution to journalConference article

4 Citations (Scopus)

Abstract

Nonogram, a popular Japanese puzzle game, is a well-known NP-complete problem. A number of approaches have been proposed and some algorithms are efficient in solving puzzles with one single solution. However, many puzzles are not limited to one ideal single solution. If there are multiple solutions to a puzzle, even the puzzle that is small in size may sometimes take a very long time to conquer. This type of problem is often observed to have sparse distribution of its colored cells. The existing efficient algorithms often gain no advantages, because the search space can stay huge and not reducible. For this reason, incorporating learning algorithms can be beneficial to support the deficiency. The objective of this study is to develop a heuristic means by the concept of simulated annealing (SA) to learn to explore different types of search spaces. Experiments are conducted to solve a number of multi-solution puzzles and the effectiveness of this approach is discussed.

Original languageEnglish (US)
Pages (from-to)541-548
Number of pages8
JournalProcedia Computer Science
Volume36
Issue numberC
DOIs
StatePublished - Jan 1 2014
EventComplex Adaptive Systems, 2014 - Philadelphia, United States
Duration: Nov 3 2014Nov 5 2014

Fingerprint

Simulated annealing
Learning algorithms
Computational complexity
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

@article{dfb511ead494410c8b4643241e2bb1a0,
title = "Simulated annealing approach to solve nonogram puzzles with multiple solutions",
abstract = "Nonogram, a popular Japanese puzzle game, is a well-known NP-complete problem. A number of approaches have been proposed and some algorithms are efficient in solving puzzles with one single solution. However, many puzzles are not limited to one ideal single solution. If there are multiple solutions to a puzzle, even the puzzle that is small in size may sometimes take a very long time to conquer. This type of problem is often observed to have sparse distribution of its colored cells. The existing efficient algorithms often gain no advantages, because the search space can stay huge and not reducible. For this reason, incorporating learning algorithms can be beneficial to support the deficiency. The objective of this study is to develop a heuristic means by the concept of simulated annealing (SA) to learn to explore different types of search spaces. Experiments are conducted to solve a number of multi-solution puzzles and the effectiveness of this approach is discussed.",
author = "Wen-li Wang and Tang, {Mei Huei}",
year = "2014",
month = "1",
day = "1",
doi = "10.1016/j.procs.2014.09.052",
language = "English (US)",
volume = "36",
pages = "541--548",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",
number = "C",

}

Simulated annealing approach to solve nonogram puzzles with multiple solutions. / Wang, Wen-li; Tang, Mei Huei.

In: Procedia Computer Science, Vol. 36, No. C, 01.01.2014, p. 541-548.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Simulated annealing approach to solve nonogram puzzles with multiple solutions

AU - Wang, Wen-li

AU - Tang, Mei Huei

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Nonogram, a popular Japanese puzzle game, is a well-known NP-complete problem. A number of approaches have been proposed and some algorithms are efficient in solving puzzles with one single solution. However, many puzzles are not limited to one ideal single solution. If there are multiple solutions to a puzzle, even the puzzle that is small in size may sometimes take a very long time to conquer. This type of problem is often observed to have sparse distribution of its colored cells. The existing efficient algorithms often gain no advantages, because the search space can stay huge and not reducible. For this reason, incorporating learning algorithms can be beneficial to support the deficiency. The objective of this study is to develop a heuristic means by the concept of simulated annealing (SA) to learn to explore different types of search spaces. Experiments are conducted to solve a number of multi-solution puzzles and the effectiveness of this approach is discussed.

AB - Nonogram, a popular Japanese puzzle game, is a well-known NP-complete problem. A number of approaches have been proposed and some algorithms are efficient in solving puzzles with one single solution. However, many puzzles are not limited to one ideal single solution. If there are multiple solutions to a puzzle, even the puzzle that is small in size may sometimes take a very long time to conquer. This type of problem is often observed to have sparse distribution of its colored cells. The existing efficient algorithms often gain no advantages, because the search space can stay huge and not reducible. For this reason, incorporating learning algorithms can be beneficial to support the deficiency. The objective of this study is to develop a heuristic means by the concept of simulated annealing (SA) to learn to explore different types of search spaces. Experiments are conducted to solve a number of multi-solution puzzles and the effectiveness of this approach is discussed.

UR - http://www.scopus.com/inward/record.url?scp=84938597128&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84938597128&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2014.09.052

DO - 10.1016/j.procs.2014.09.052

M3 - Conference article

AN - SCOPUS:84938597128

VL - 36

SP - 541

EP - 548

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

IS - C

ER -