350 rub
Journal Neurocomputers №5 for 2013 г.
Article in number:
Functional pre-treatment of input signals in a neural network
Keywords: neural network neural network modelling correlation neural network package
Authors:
A.L. Gusev, F.M. Cherepanov, L.N. Yasnitsky
Abstract:
In this paper, the authors propose a technique to intensify the correlations between input and output signals within a neural network by means of the adjustment of input signals. This adjustment is performed using specially selected elementary functions for each input of a neural network. The presented technique improves the accuracy of neural network models in modelling complicated processes. The authors suggest embedding the proposed algorithm into neural network packages to improve the quality of neural network modelling.
Pages: 19-21
References

  1. Gorban A. N.Nejjroinformatika i ee prilozhenija // Otkrytye sistemy. 1998. № 04 - 06.
  2. Gorban A. N., Mirkes E. M. Nejjronnye seti associativnojj pamjati, funkcionirujushhie v diskretnom vremeni / Vychislitelnyjj centr SO RAN. Krasnojarsk. Rukopis dep. V VINITI 17.07.97. №2436-V97.
  3. Kruglov V. V., Borisov V. V. Iskusstvennye nejjronnye seti. Teorija i praktika. M.: Gorjachaja linija - Telekom. 2002.
  4. Caregorodcev V. G. Optimizacija predobrabotki dannykh: konstanta Lipshica obuchajushhejj vyborki i svojjstva obuchennykh nejjronnykh setejj // Nejjrokompjutery: razrabotka, primenenie. 2003. № 7. S. 3-8.
  5. CHerepanov F. M., JAsnickijj L. N. Nejjrosetevojj filtr dlja iskljuchenija vybrosov v statisticheskojj informacii // Vestnik Permskogo universiteta. Matematika. Mekhanika. Informatika. 2008. Vyp. 4 (20). S. 151-155.
  6. CHerepanov F. M., JAsnickijj L. N. Simuljator nejjronnykh setejj «Nejjrosimuljator 1.0» // Svidetelstvo ob otraslevojj registracii razrabotki № 8756. Zaregistrirovano v Otraslevom fonde algoritmov i programm 12.07.2007.
  7. JAsnickijj L. N., Zajjceva N. V., Gusev A. L., SHur P. Z. Nejjrosetevaja model regiona dlja vybora upravljajushhikh vozdejjstvijj v oblasti obespechenija gigienicheskojj bezopasnosti // Informatika i sistemy upravlenija. 2011. №3 (29). S. 51-59.
  8. JAsnickijj L. N., Sichinava Z. I.Nejjrosetevye algoritmy analiza povedenija respondentov // Nejjrokompjutery: razrabotka, primenenie. 2011. № 10. S. 59-64.
  9. JAsnickijj L. N., CHerepanov F. M. O vozmozhnostjakh primenenija nejjrosetevykh tekhnologijj v politologii // Nejjrokompjutery: razrabotka, primenenie. 2010. № 8. Vyp. 4. S. 47-53.
  10. Pèrez-Ilzarbe, M. J.,Preconditioning method to accelerate neural networks gradient training algorithms / Proc. IJCNN, Washington, DC, USA, 1999.
  11. Schraudolph N. N. Centering neural network gradient factors / Neural networks: Tricks of the trade. Springer Verlag. Lecture Notes in Comp. Sci. 1998. V. 1524. P.207-226.