Geodesic
Approaches to solving systems of linear algebraic equations using neural networks
Vestnik KRAUNC. Fiziko-matematičeskie nauki, Tome 40 (2022) no. 3, pp. 153-164 Cet article a éte moissonné depuis la source Math-Net.Ru

Voir la notice de l'article

System linear is the main solution for an essential class of mathematical modeling problems. The study of the possibility of solving system linear using neural networks will allow creating new approaches to solving problems of mathematical modeling. A new way of solving systems of linear equations using neural networks is presented. Feedforward networks and a stochastic gradient descent algorithm are used. The stages of designing a neural network are described, as well as the process of choosing the optimal NN structure, based on the computational experiments performed. The results of using neural networks for solving systems of linear equations are presented. The expediency of using NN for problems of this type is substantiated.
Keywords: systems of linear algebraic equations, Neural networks
Mots-clés : gradient descent. 
@article{VKAM_2022_40_3_a12,
     author = {V. A. Galkin and T. V. Gavrilenko and A. D. Smorodinov},
     title = {Approaches to solving systems of linear algebraic equations using neural networks},
     journal = {Vestnik KRAUNC. Fiziko-matemati\v{c}eskie nauki},
     pages = {153--164},
     year = {2022},
     volume = {40},
     number = {3},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VKAM_2022_40_3_a12/}
}
TY  - JOUR
AU  - V. A. Galkin
AU  - T. V. Gavrilenko
AU  - A. D. Smorodinov
TI  - Approaches to solving systems of linear algebraic equations using neural networks
JO  - Vestnik KRAUNC. Fiziko-matematičeskie nauki
PY  - 2022
SP  - 153
EP  - 164
VL  - 40
IS  - 3
UR  - http://geodesic.mathdoc.fr/item/VKAM_2022_40_3_a12/
LA  - ru
ID  - VKAM_2022_40_3_a12
ER  - 
%0 Journal Article
%A V. A. Galkin
%A T. V. Gavrilenko
%A A. D. Smorodinov
%T Approaches to solving systems of linear algebraic equations using neural networks
%J Vestnik KRAUNC. Fiziko-matematičeskie nauki
%D 2022
%P 153-164
%V 40
%N 3
%U http://geodesic.mathdoc.fr/item/VKAM_2022_40_3_a12/
%G ru
%F VKAM_2022_40_3_a12
V. A. Galkin; T. V. Gavrilenko; A. D. Smorodinov. Approaches to solving systems of linear algebraic equations using neural networks. Vestnik KRAUNC. Fiziko-matematičeskie nauki, Tome 40 (2022) no. 3, pp. 153-164. http://geodesic.mathdoc.fr/item/VKAM_2022_40_3_a12/

[1] Manzhula V. G., Fedyashov D. S., “Neironnye seti Kokhonena i nechetkie neironnye seti v intellektualnom analize dannykh”, Fundamentalnye issledovaniya, 2011, no. 4, 108-114

[2] Dolotov E. A. Kustikova V. D., “Sravnenie nekotorykh metodov resheniya zadachi detektirovaniya lits na izobrazheniyakh”, GraphiCon, 2017, no. 2017, 202-207

[3] Korneev D. S., “Ispolzovanie apparata neironnykh setei dlya sozdaniya modeli otsenki i upravleniya riskami predpriyatiya”, Upravlenie bolshimi sistemami, 2007, no. 17, 81-102

[4] Fransua Sh., Glubokoe obuchenie na Python, Piter, SPb, 2018, 400 pp.

[5] Popova Yu. B., Yatsynovich S. V., Obuchenie iskusstvennykh neironnykh setei metodom obratnogo rasprostraneniya oshibki, BNTU, Minsk, 2016

[6] Ivanovskii M. N., Shafeev O. P., “Primenenie metoda obratnogo rasprostraneniya oshibki dlya obucheniya neironnoi seti”, Informatsionnye tekhnologii v nauke i proizvodstve, V Vserossiiskaya molodezhnaya nauchno-tekhnicheskaya konferentsiya, 2018, 39–43

[7] Utkin P. S., Lektsiya po teme: Vvedenie v reshenie SLAU Normy vektorov i matrits. Chislo obuslovlennosti matritsy SLAU, MFTI, M., 2014

[8] Cybenko, G. V., “Approximation by Superpositions of a Sigmoidal function”, Mathematics of Control Signals and Systems, 2 (1989), 303–314 | DOI

[9] Shiyu Liang, R. Srikant, “Why deep neural networks for function approximation?”, Published as a conference paper at ICLR, 2017

[10] Hanin B., “Universal Function Approximation by Deep Neural Nets with Bounded Width and ReLU Activations”, Mathematics, 7 (2019), 992 | DOI