Geodesic
Initialization of multilayered forecasting artificial neural networks
Učënye zapiski Kazanskogo universiteta. Seriâ Fiziko-matematičeskie nauki, Uchenye Zapiski Kazanskogo Universiteta. Seriya Fiziko-Matematicheskie Nauki, Tome 152 (2010) no. 1, pp. 7-14 Cet article a éte moissonné depuis la source Math-Net.Ru

Voir la notice du chapitre de livre

In this paper a method of initializing a neural network solving the problem of forecasting a time series is offered. The analogy with a linear prediction filter is used to select an initial weighting coefficient for neurons. Also the variants of decomposition of a transformation matrix corresponding to the linear prediction filter are offered to improve the initialization method quality. Through the neural nets forecasting of the Lorentz chaotic system's trajectory it is shown that the application of the given method allows significantly increasing the accuracy of the neural network prediction.
Keywords: neurofuzzy modeling, forecasting, neural network initialization, linear prediction filter. 
@article{UZKU_2010_152_1_a0,
     author = {V. V. Bochkarev and Yu. S. Maslennikova},
     title = {Initialization of multilayered forecasting artificial neural networks},
     journal = {U\v{c}\"enye zapiski Kazanskogo universiteta. Seri\^a Fiziko-matemati\v{c}eskie nauki},
     pages = {7--14},
     year = {2010},
     volume = {152},
     number = {1},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/UZKU_2010_152_1_a0/}
}
TY  - JOUR
AU  - V. V. Bochkarev
AU  - Yu. S. Maslennikova
TI  - Initialization of multilayered forecasting artificial neural networks
JO  - Učënye zapiski Kazanskogo universiteta. Seriâ Fiziko-matematičeskie nauki
PY  - 2010
SP  - 7
EP  - 14
VL  - 152
IS  - 1
UR  - http://geodesic.mathdoc.fr/item/UZKU_2010_152_1_a0/
LA  - ru
ID  - UZKU_2010_152_1_a0
ER  - 
%0 Journal Article
%A V. V. Bochkarev
%A Yu. S. Maslennikova
%T Initialization of multilayered forecasting artificial neural networks
%J Učënye zapiski Kazanskogo universiteta. Seriâ Fiziko-matematičeskie nauki
%D 2010
%P 7-14
%V 152
%N 1
%U http://geodesic.mathdoc.fr/item/UZKU_2010_152_1_a0/
%G ru
%F UZKU_2010_152_1_a0
V. V. Bochkarev; Yu. S. Maslennikova. Initialization of multilayered forecasting artificial neural networks. Učënye zapiski Kazanskogo universiteta. Seriâ Fiziko-matematičeskie nauki, Uchenye Zapiski Kazanskogo Universiteta. Seriya Fiziko-Matematicheskie Nauki, Tome 152 (2010) no. 1, pp. 7-14. http://geodesic.mathdoc.fr/item/UZKU_2010_152_1_a0/

[1] Anderson T., Statisticheskii analiz vremennykh ryadov, Mir, M., 1976, 757 pp.

[2] Arato M., Lineinye stokhasticheskie sistemy s postoyannymi koeffitsientami. Statisticheskii podkhod, Nauka, M., 1989, 304 pp. | MR | Zbl

[3] Galushkin A. I., Teoriya neironnykh setei, IPRZhR, M., 2000, 416 pp.

[4] Kruglov V. V., Dli M. I., Golunov R. Yu., Nechëtkaya logika i iskusstvennye neironnye seti, Izd-vo fiz.-matem. lit., M., 2001, 224 pp.

[5] Medvedev V. S., Potëmkin V. G., Neironnye seti. MATLAB 6, DIALOG-MIFI, M., 2002, 496 pp.

[6] Gholipour A., Lucas C., Araabi B. N., Shafiee M., “Solar activity forecast: Spectral analysis and neurofuzzy prediction”, J. Atmos. Solar-Terrestr. Phys., 67 (2005), 595–603 | DOI

[7] Gril F., Myurrei U., Rait M., Prakticheskaya optimizatsiya, Mir, M., 1985, 509 pp. | MR

[8] Kronover R. M., Fraktaly i khaos v dinamicheskikh sistemakh. Osnovy teorii, Postmarket, M., 2000, 352 pp.

[9] Frik P. G., Turbulentnost: podkhody i metody, IKI, Moskva–Izhevsk, 2003, 292 pp.

[10] Pu T., Nelineinaya ekonomicheskaya dinamika, NITs “RKhD”, Moskva–Izhevsk, 2000, 198 pp.