Article 2215

Title of the article

TREES GENERATING ALGORITHMS FOR GENETIC SEARCH

Authors

Makarychev Petr Petrovich, Doctor of engineering sciences, professor, head of sub-department of computer
application and software, Penza State University (40 Krasnaya street, Penza, Russia), makpp@yandex.ru
Sleptsov Nikolay Vladimirovich, Candidate of engineering sciences, associate professor, sub-department of management, Penza State University (40 Krasnaya street, Penza, Russia), nbs_nbs@km.ru

Index UDK

681.3.01:681.3.05

Abstract

Background. Application of evolutionary computing methods effectively to solve a wide range subformulae tasks. Еvolutionary computations require complex solu-tions of a number of problems that include generation of solution structures. This fact proves relevance of the issues under consideration.
Materials and methods. Genetic programming is an evolutionary method of optimi-zation that generates functional programs to solve specific tasks. Usually the programs are shaped in the form of a tree, which is interpreted as the s-expression in LISP lan-guage. The programs/trees mass generation during each cycle of evolutionary modeling makes quite high demands to the quality of the generating algorithms on speed, tree size and probabilistic characteristics of nodes. The proposed algorithms combine prop-erties of highly efficient solutions, from the point of view of structures’ generation, en-suring high stability of genetic modeling.
Results and conclusions. The authors have proposed new algorithms that gener-ate uniformly distributed structures and lower computational complexity and man-agement of their characteristics based on user data.

Key words

genetic programming, population, tree generation, mutation, tree growth.

Download PDF
References

1. Engel'gardt V. V. Izvestiya vysshikh uchebnykh zavedeniy. Povolzhskiy region. Tekhnicheskie nauki [University proceedings. Volga region. Engineering sciences]. 2013, no. 4 (28), pp. 5–15.
2. Chellapilla K. Genetic Programming 1997: Proceedings of the Second Annual Confer-ence (GP97). San Fransisco: Morgan Kaufmann, 1997, pp. 431–438.
3. Luke S. A and Spector L. Genetic Programming 1998: Proceedings of the Third Annual Conference (GP98). San Fransisco, CA: Morgan Kaufmann, 1998, pp. 208–213.
4. Luke S. ECJ: A Java-based evolutionary computation and genetic programming sys-tem, 2000. Available at: http://www.cs.umd.edu/projects/plus/ecj/

 

Дата создания: 02.10.2015 15:10
Дата обновления: 02.10.2015 15:51