Geodesic
A parallel algorithm of the multivariant evolutionary synthesis of nonlinear models
Sibirskij žurnal vyčislitelʹnoj matematiki, Tome 20 (2017) no. 2, pp. 169-180 
O. G. Monakhov; E. A. Monakhova. A parallel algorithm of the multivariant evolutionary synthesis of nonlinear models. Sibirskij žurnal vyčislitelʹnoj matematiki, Tome 20 (2017) no. 2, pp. 169-180. http://geodesic.mathdoc.fr/item/SJVM_2017_20_2_a4/
@article{SJVM_2017_20_2_a4,
     author = {O. G. Monakhov and E. A. Monakhova},
     title = {A parallel algorithm of the multivariant evolutionary synthesis of nonlinear models},
     journal = {Sibirskij \v{z}urnal vy\v{c}islitelʹnoj matematiki},
     pages = {169--180},
     year = {2017},
     volume = {20},
     number = {2},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/SJVM_2017_20_2_a4/}
}
TY  - JOUR
AU  - O. G. Monakhov
AU  - E. A. Monakhova
TI  - A parallel algorithm of the multivariant evolutionary synthesis of nonlinear models
JO  - Sibirskij žurnal vyčislitelʹnoj matematiki
PY  - 2017
SP  - 169
EP  - 180
VL  - 20
IS  - 2
UR  - http://geodesic.mathdoc.fr/item/SJVM_2017_20_2_a4/
LA  - ru
ID  - SJVM_2017_20_2_a4
ER  - 
%0 Journal Article
%A O. G. Monakhov
%A E. A. Monakhova
%T A parallel algorithm of the multivariant evolutionary synthesis of nonlinear models
%J Sibirskij žurnal vyčislitelʹnoj matematiki
%D 2017
%P 169-180
%V 20
%N 2
%U http://geodesic.mathdoc.fr/item/SJVM_2017_20_2_a4/
%G ru
%F SJVM_2017_20_2_a4

Voir la notice de l'article provenant de la source Math-Net.Ru

A parallel algorithm for solving the problem of constructing of nonlinear models (mathematical expressions, functions, algorithms, programs) based on given experimental data, a set of variables, basic functions and operations is proposed. The proposed algorithm of the multivariant evolutionary synthesis of nonlinear models has a linear representation of the chromosome, the modular operations in decoding the genotype to the phenotype for interpreting a chromosome as a sequence of instructions, the multivariant method for presenting a multiplicity of models (expressions) using a single chromosome. A comparison of the sequential version of the algorithm with a standard algorithm of genetic programming and the algorithm of the Cartesian Genetic Programming offers advantage of the algorithm proposed both in the time of obtaining a solution (by about an order of magnitude in most cases), and in the probability of finding a given function (model). In the experiments on the parallel supercomputer systems, estimates of the efficiency of the proposed parallel algorithm have been obtained showing linear acceleration and scalability.

[1] Koza J, Genetic Programming, v. II, Automatic Discovery of Reusable Programs, MIT Press, Cambridge, 1996

[2] Langdon W. B., Poli R., Foundations of Genetic Programming, Springer-Verlag, 2002 | Zbl

[3] Poli R., Langdon W. B., McPhee N. F., A Field Guide to Genetic Programming, Lulu.com, San Francisco, California, USA, 2008

[4] Emelyanov V. V., Kureichik V. V., Kureichik V. M., Teoriya i praktika evolyutsionnogo modelirovaniya, FIZMATLIT, M., 2003

[5] Monakhov O. G., “Issledovanie vliyaniya stepeni spetsializatsii shablonov na prostranstvo poiska pri evolyutsionnom sinteze modelei”, Prikladnaya diskretnaya matematika, 2012, no. 3, 85–95

[6] Miller J. F., Cartesian Genetic Programming, Springer, 2011 | Zbl

[7] Monakhova E. A., Monakhov O. G., “Poisk rekordnykh tsirkulyantnykh grafov s ispolzovaniem parallelnogo geneticheskogo algoritma”, Diskretnyi analiz i issledovanie operatsii, 22:6 (2015), 29–42 | DOI | MR | Zbl

[8] Oltean M., Multi Expression Programming, Technical Report, Babes-Bolyai Univ., Romania, 2006