Article 12319

Title of the article

PROVIDING COMPUTATIONAL EFFICIENCY OF GENETIC SEARCH 

Authors

Imamutdinov Anton Igorevich, postgraduate student, Penza State University (440026, 40 Krasnaya street, Penza, Russia), E-mail: antonim94@yandex.ru 

Index UDK

519.718 

DOI

10.21685/2307-4205-2019-3-12 

Abstract

Background. Evaluation of reliability and quality indicators of multiparameter dependencies is possible by means of evolutionary-genetic calculations that do not require data on the internal structure of the connections and dependencies of the analyzed objects. Evolutionary methods show high performance, either with the application of specialized types of genetic search, or with effective management of the search in the solution process.
Methods. A promising method for increasing the efficiency of optimization models of computations based on evolutionary methods with the universality of the computation model is to preserve arbitrary lengths in the encoding of indestructible information, which places additional demands on the implementation of the main genetic operators.
Results. Additional requirements can be reduced to the properties of preserving templates with high rates regardless of their length, as well as the ability to highlight, determine their properties, note the suitability. The implementation of such an extension in the search management is proposed to implement the introduction of masks and mask control mechanism.
Conclusions. The proposed approach to improving the efficiency of genetic search can be implemented using modified genetic operations based on masking and a fixed set of rules to significantly increase computational costs. 

Key words

optimization, evolutionary search models, genetic operators, masking functions 

Download PDF

 

Дата создания: 01.11.2019 13:04
Дата обновления: 01.11.2019 14:09