An algorithm for evolutionary synthesis is presented. It combines the advantages of geneticalgorithms and genetic programming based on evolutionary computation, template (pattern,skeleton) of algorithm and a given set of pairs of input-output data. The influence oftemplate specialization degree on the search space in the evolutionary synthesis is investigated.Estimates are obtained for the reduction of the search space with the introducingan additional information of template in the form of formulas represented by binary treesand refining the model. The exponential and superexponential reduction in the time of thesynthesis algorithm with the increasing the template specialization degree is shown theoreticallyand experimentally.
Download file
Counter downloads: 67
- Title Influence of template specialization degree on the search space inevolutionary synthesis of models
- Headline Influence of template specialization degree on the search space inevolutionary synthesis of models
- Publesher
Tomsk State University - Issue Prikladnaya Diskretnaya Matematika - Applied Discrete Mathematics 3(17)
- Date:
- DOI
Keywords
генетический алгоритм, генетическое программирование, эволюционный синтез, шаблон, темплейт, скелетон, genetic algorithms, genetic programming, evolutionary synthesis, template, design patternAuthors
References
http: //oeis.org - On-Line Encyclopedia of Integer Sequences.
Koshy T. Catalan numbers with applications. Oxford: Oxford University Press, 2009. 422 p.
Sloane N. J. A. and PlouffeS. The Encyclopedia of Integer Sequences. London: Academic Press, 1995. 587 p.
Koza J. Genetic Programming. Cambridge: The MIT Press, 1992. 814 p.
Monakhova E., Monakhov O., and Mukhoed E. Genetic Construction of Optimal Circulant Network Designs / / LNCS. 1999. V. 1596. P. 215-223.
Goldberg D. E. Genetic Algorithms, in Search, Optimization and Machine Learning. Reading, MA.: Addison-Wesley, 1989. 432 p.
Монахов О. Г. Программа эволюционного поиска математических описаний недоопределенных функций на основе шаблонов / / Свид. об официальной регистрации программы на ЭВМ №2011611546. М.: Федеральная служба по интеллектуальной собственности, патентам и товарным знакам, 2011.
Monakhov O. and Monakhova E. An Algorithm for Discovery of New Families of Optimal Regular Networks / / Proc. 6th Inter. Conf. on Discovery Science (DS 2003), Oct. 17-20, 2003, Sapporo, Japan. Lecture Notes in Artificial Intelligence. 2003. V. 2843. P. 244-254.
Монахов О. Г., Монахова Э. А. Синтез новых семейств оптимальных регулярных сетей на основе эволюционных вычислений и шаблонов функций / / Автометрия. 2004. №4. С.106-116.
Монахов О. Г. Эволюционный синтез алгоритмов на основе шаблонов / / Автометрия. 2006. №1. С. 116-126.
Gamma E., Helm R., Johnson R., and Vlissides J. Design Patterns: Elements of Reusable Object-Oriented Software. Reading, MA.: Addison-Wesley, 1994. 416 p.
Watanobe Y., Mirenkov N., Yoshioka R., and Monakhov O. Filmification of methods: A visual language for graph algorithms / / J. Visual Languages and Computing. 2008. No. 1. P. 123-150.
Mirenkov N. and Mirenkova T. Multimedia Skeletons and "Filmification" of Methods / / Proc. First Intern. Conf. Visual Information Systems. Melbourne, Australia: Victoria University, 1996. P. 58-67.
Cole M. Algorithmic Skeletons: Structured Management of Parallel Computation. Cambridge: The MIT Press, 1989. 131 p.
Influence of template specialization degree on the search space inevolutionary synthesis of models | Prikladnaya Diskretnaya Matematika - Applied Discrete Mathematics. 2012. № 3(17).
Download full-text version
Download fileCounter downloads: 248