Automated, adaptive, and optimized search for CSPs via cuckoo search

RICARDO JAVIER SOTO DE GIORGIS, BRODERICK CRAWFORD LABRIN, Javier Flores, Felipe Mella, Cristian Galleguillos, Franklin Johnson, Fernando Paredes

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Constraint Programing is a programming paradigm devoted to the efficient solving of constraint satisfaction problems (CSPs). A CSP is a formal problem representation mainly composed of variables and constraints defining relations among those variables. The resolution process of CSPs is commonly carried out by building and exploring a search tree that holds the possibles solutions. Such a tree is dynamically created by interleaving two different phases: enumeration and propagation. During enumeration, the variables and values are chosen to build the possible solution, while propagation intend to delete the values having no chance to reach a feasible result. Autonomous Search is a new technique that gives the ability to the resolution process to be adaptive by re-configuring its enumeration strategy when poor performances are detected. This technique has exhibited impressive results during the last years. However, such a re-configuration is hard to achieve as parameters are problem-dependent and their best configuration is not stable along the search. In this paper, we introduce an Autonomous Search framework that incorporates a new optimizer based on Cuckoo Search able to efficiently support the re-configuration phase. Our goal is to provide an automated, adaptive, and optimized search system for CSPs. We report encouraging results where our approach clearly improves the performance of previously reported Autonomous Search approaches for CSPs.

Original languageEnglish
Title of host publicationAdvances in Swarm and Computational Intelligence - 6th International Conference, ICSI 2015 held in conjunction with the 2nd BRICS Congress, CCI 2015, Proceedings
EditorsAlexander Gelbukh, Ying Tan, Swagatam Das, Andries Engelbrecht, Fernando Buarque, Yuhui Shi
PublisherSpringer Verlag
Pages436-447
Number of pages12
ISBN (Print)9783319204659
DOIs
StatePublished - 1 Jan 2015
Event6th International Conference on Swarm Intelligence, ICSI 2015 held in conjunction with the 2nd BRICS Congress on Computational Intelligence, CCI 2015 - Beijing, China
Duration: 25 Jun 201528 Jun 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9140
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th International Conference on Swarm Intelligence, ICSI 2015 held in conjunction with the 2nd BRICS Congress on Computational Intelligence, CCI 2015
CountryChina
CityBeijing
Period25/06/1528/06/15

Keywords

  • Autonomous search
  • Constraint programming
  • Cuckoo search
  • Nature-inspired algorithms
  • Swarm-based optimization

Fingerprint Dive into the research topics of 'Automated, adaptive, and optimized search for CSPs via cuckoo search'. Together they form a unique fingerprint.

Cite this