INVESTIGATION OF THE IMPACT OF POPULATION SIZE ON THE PERFORMANCE OF A GENETIC ALGORITHM
Abstract
The paper investigates ways to determine the population size in a genetic algorithm and studies the
relationship between the number of individuals and the speed of the algorithm. Methods for determining
the optimal number of individuals in a population by different methods are described: depending on the
size of the chromosomes, for a tree-like type of chromosomes, in the presence of a noise factor and by the
method of a neighboring element with a maximum and minimum boundary. The data obtained by performing
each method differ from each other, for this reason, an assessment was made in order to verify the
accuracy of theoretical data by comparing them with experimental ones. To conduct experiments, a program was developed on the Unity graphics platform with the ability to change the number of individuals in
the population. After receiving the results, the experimental data were compared with the data obtained on
the basis of methods for determining the population size in the genetic algorithm from the first part of the
work. The experiment showed that the optimal population size lies in the range of 100-160 individuals.
With a decrease in their number, the execution time of the task begins to increase significantly, and with
an increase beyond the calculated limit, the reduction in execution time does not correspond to the computing
resources expended. The experimental data obtained themselves have the smallest error with the
method used by the tree representation of chromosomes. The results of the study can be used to select the
size of the population during training in order to achieve a better ratio of computing power to learning
speed, and a method defined in the course of work can help in theoretical calculations
References
sistem pravil [The use of genetic algorithms in data analysis tasks and the construction of rule systems],
Novye informatsionnye tekhnologii v avtomatizirovannykh sistemakh [New information technologies
in automated systems], 2013, No. 16, pp. 72-77.
2. Shchukin M.A. Geneticheskie algoritmy kak metod resheniya matematicheskikh zadach [Genetic algorithms
as a method of solving mathematical problems], Vestnik URAO [Bulletin of the University of
the Russian Academy of Education], 2014, No. 5, pp. 168-172.
3. Korneev A.S., Lyakhov M.S., Tarnetskaya A.V. Ispol'zovanie geneticheskikh algoritmov dlya resheniya
zadach optimizatsii v elektroenergetike [The use of genetic algorithms for solving optimization problems
in the electric power industry], IX Vserossiyskaya nauchno-prakticheskaya konferentsiya
molodykh uchenykh «ROSSIYA MOLODAYA». Aprel' 2017 [IX All-Russian Scientific and Practical
Conference of Young Scientists “YOUNG RUSSIA”. April 2017], 4 p. Available at:
https://science.kuzstu.ru/wp-content/Events/Conference/RM/2017/RM17/pages/Articles/0201038.
4. Shcheglov S.N. Analiz modeley prinyatiya resheniy v novykh informatsionnykh tekhnologiyakh v
usloviyakh nechetkosti i neopredelennosti [Analysis of decision-making models in new information
technologies in conditions of fuzziness and uncertainty], Izvestiya YuFU. Tekhnicheskie nauki
[Izvestiya SFedU. Engineering Sciences], 2012, No. 132 (7), pp. 126-131.
5. Kureychik V.V., Kureychik Vl.Vl, Saak A.E. Evolyutsionnyy algoritm dlya resheniya zadachi
dispetcherizatsii [An evolutionary algorithm for solving the dispatching problem. Iyul], Izvestiya
YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2021, No. 2 (219). pp. 50-59.
6. Janne Koljonen, Jarmo Alander. Effects of population size and relative elitism on optimization speed
and reliability of genetic algorithms. January 2006, pp. 1-7.
7. Yu-an Zhang, Makoto Sakamoto, Hiroshi Furutani. Effects of Population Size and Mutation Rate on
Results of Genetic Algorithm, January 2008, pp. 47-55.
8. Tsoy Yu.R. Vliyanie razmera populyatsii i poiska ot ogranicheniya po vremeni raboty geneticheskogo
algoritma [The influence of population size and search time limit on genetic algorithm]. Tomsk:
Tomskiy politekhnicheskiy universitet, 2003, 7 p.
9. Goncharova A.S., Beketova O.S., Zarosilo L.R. Modeli geneticheskikh algoritmov [Models of genetic
algorithms], Molodoy uchenyy [Young scientist]. Fevral' 2015, No. 1 (1), pp. 32-33.
10. Tsoy Yu.R. Reshenie zadach lineynogo programmirovaniya s pomoshch'yu geneticheskogo algoritma
[Solving linear programming problems using a genetic algorithm]. Tomsk: Tomskiy politekhnicheskiy
universitet, 2003, 2 p.
11. Chipiga A.F., Petrov Yu.Yu. Otsenka vybora razmera populyatsii v prostom geneticheskom algoritme
[Estimating the choice of population size in a simple genetic algorithm], Informatsionnoe
protivodeystvie ugrozam [Information counteraction to threats], 2005, No. 5, pp. 26-29.
12. Glebov A.V., Strikelev D.A. Vliyanie razmera populyatsii na skorost' skhodimosti geneticheskogo
algoritma pri drevovidnom predstavlenii khromosom [The influence of population size on the convergence
rate of a genetic algorithm with a tree-like representation of chromosomes], Elektronnaya
biblioteka BGU [Electronic Library of the BSU]. Aprel' 2010, pp. 3. Available at:
https://elib.bsu.by/handle/123456789/28657.
13. Glebov A., Strykeleu D. Experimental convergence study of a genetic algorithm using
consistencypreserving genetic operators for tree-like chromosomes, Modern problems of mathematics
and computer engineering, 2009, pp. 3-5.
14. Randy Haupt. Optimum Population Size and Mutation Rate for a Simple Real Genetic Algorithm that Optimizes
Array Factors, Antennas and Propagation Society International Symposium, 2000, pp. 173-186.
15. Harik G.R., Cantu-Paz E., Goldberg D.E., Miller B.L. The Gambler's Ruin Problem, Genetic Algorithms,
and the Sizing of Populations, Association for Computing Machinery, February 1999, pp. 231-253.
16. Frolov V.V. Opredelenie razmera populyatsii geneticheskogo algoritma dlya zadach diskretnoy
optimizatsii v SAPR [Determining the population size of a genetic algorithm for discrete optimization
problems in CAD], Sistemnyy analiz i prikladnaya informatika [System analysis and applied informatics],
2018, No. 3, pp. 9-16.
17. Saprykin A.N., Akinina K.D., Saprykina E.N. Optimal'noe chislo konvergiruemykh pokoleniy i
poleznykh osobey pri var'iruyushchikhsya znacheniyakh veroyatnosti mutatsii geneticheskogo
algoritma na malykh prostranstvakh poiska [The optimal number of convergent generations and useful
individuals with varying values of the mutation probability of a genetic algorithm in small search
spaces] Sovremennye tekhnologii v nauke i obrazovanii: Sb. trudov II mezhdunarodnoy nauchnotekhnicheskoy
i nauchno-metodicheskoy konferentsii [Modern technologies in science and education:
Collection of proceedings of the II international scientific-technical and scientific-methodological conference].
Vol. 4. Ryazan': RRTI, 2017, pp. 248-252.
18. Roeva O., Fidanova S., Paprzycki M. Influence of the population size on the genetic algorithm performance
in case of cultivation process modeling, Computer Science and Information Systems, 2013,
pp. 371-376.
19. Chastikova V.A. Identifikatsiya mekhanizmov realizatsii operatorov geneticheskogo algoritma v
ekspertnykh sistemakh produktsionnogo tipa [Identification of mechanisms for implementing genetic
algorithm operators in production-type expert systems], Nauchnyy zhurnal KubGAU [Scientific journal
of KubSAU], 2012, No. 75.
20. Kornilov A. UNITY. Polnoe rukovodstvo [UNITY. Complete guide]. 2nd ed. Saint Petersburg: Nauka
i Tekhnika, 2021, 496 p.
21. Tsygankov V.A., Shabalina O.A., Kataev A.V. Ispol'zovanie geneticheskikh algoritmov dlya povysheniya
skorosti obucheniya neyronnykh setey [Using genetic algorithms to increase the speed of learning of neural
networks], Inzhenernyy vestnik Dona [Engineering Bulletin of the Don], 2023, No. 3.
