Number of the records: 1
A New Approach to Tuning Heuristic Parameters of Genetic Algorithms
- 1.
SYSNO ASEP 0031808 Document Type J - Journal Article R&D Document Type Journal Article Subsidiary J Článek ve SCOPUS Title A New Approach to Tuning Heuristic Parameters of Genetic Algorithms Title Nový přístup k nastavování heuristických parametrů genetických algoritmů Author(s) Holeňa, Martin (UIVT-O) SAI, RID Source Title WSEAS Transactions on Information Science and Applications - ISSN 1790-0832
Roč. 3, č. 3 (2006), s. 562-569Number of pages 8 s. Action AIKED'06. WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases Event date 15.02.2006-17.02.2006 VEvent location Madrid Country ES - Spain Event type WRD Language eng - English Country US - United States Keywords evolutionary optimization ; genetic algorithms ; heuristic parameters ; parameter tuning ; artificial neural networks ; convergence speed ; population diversity Subject RIV IN - Informatics, Computer Science R&D Projects GA201/05/0325 GA ČR - Czech Science Foundation (CSF) GA201/05/0557 GA ČR - Czech Science Foundation (CSF) CEZ AV0Z10300504 - UIVT-O (2005-2011) EID SCOPUS 33645149584 Annotation The heuristic inspiration of genetic algorithms entails their dependence on many heuristic parameters. For the algorithm to work properly, those parameters need to be empirically tuned. If the values of the objective function have to be obtained in a costly experimental way, then the algorithm can not be run with several various combinations of the values of heuristic parameters. This paper suggests to use a neural-network based approximation of the objective function for parameter tuning in such situations. Such an aproach allows to investigate the convergence speed of the algorithm and the diversity of the population for many combinations of the values of heuristic parameters. A case study in materials science demonstrates the feasibility of the proposed approach. Workplace Institute of Computer Science Contact Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Year of Publishing 2007
Number of the records: 1