Theorem of training for a competition algorithm

V. S. Antyufeev

Institute of Computational Mathematics and Mathematical Geophysics (Computing Center), Siberian Branch of the Russian Academy of Sciences, Novosibirsk

Abstract: This paper is an extension of [1], where a new decision algorithm was proposed. In its operation, the unit resembles artificial neural networks. However the functioning of the algorithm proposed is based on the different concepts. It does not use the concept of a net, a neuron. The theorem of training for the new competition algorithm is proved.

Key words: theorem of training, probabilistic convergence, artificial neural network.

UDC: 519.633

Received: 17.11.2011
Revised: 16.12.2011

 English version: Numerical Analysis and Applications, 2013, 6:1, 1–8

Bibliographic databases:

• 
• 
•