Russian Federation
The possibilities of using genetic algorithms (GA) as a basis for creating artificial intelligence (AI) systems are considered. The advantages of GA include its versatility, which allows it to be used for solving various problems that require finding the optimal solution, as well as its ability to work with incomplete or noisy data.The article provides a detailed discussion of the features of finding the optimal solution based on the evolutionary process and the GA algorithm, which includes the main steps of GA operation: creating an initial population, selecting the most fit individuals based on the fitness function (objective function), crossing (recombination), and mutation. The scaling of the fitness function (objective function) is considered, performed by one of three methods: linear scaling, power scaling, and sigma-clipping. The methods of encoding information in GA are considered in detail, including integer, real, and logarithmic encoding.
artificial intelligence, genetic algorithm, information encoding, scaling, mutation, population, recombination, selection, fitness function, objective function, optimization
1. Averkin A.N., Gaaze-Rapoport M.G., Pospelov D.A. Explanatory Dictionary on Artificial Intelligence. M.: Radio and Communication, 1992. 256 p.
2. Devyatkov V. V. Artificial Intelligence Systems. M.: Publishing House of the Bauman Moscow State Technical University, 2001.
3. Gladkov L. A., Kureichik V. V., Kureichik V. M., et al. Bio-inspired Methods in Optimization: A Monograph. Moscow: Fizmatlit, 2009.
4. Baluja S. Genetic algorithms and search statistics. MIT Pres
5. Davis L. Handbook of Genetic Algorithms. “Van Nostrand Reinhold”, 2009.
6. Michalewitch Z. Genetic Algorithms. Springer-Verlag, 2012.
7. Labinsky A.Yu., Shcherbakov O.V. Features of the use of computer simulation of evolutionary processes // Problems of risk management in the technosphere. 2017.
8. Labinsky A.Yu. Use of a genetic algorithm for multi-criteria optimization // Problems of risk management in the technosphere. No. 4, 2018.
9. Labinsky A.Yu. Multiparametric optimization using a genetic algorithm // Problems of risk management in the technosphere. No. 2, 2020.
10. Labinsky A.Y. Promising areas of computer modeling of complex processes and systems: monograph. St. Petersburg: SPbU GPS EMERCOM of Russia, 2017.
