Discrete Dynamics in Nature and Society 
Volume 3 (1999), Issue 1, Pages 43-49
doi:10.1155/S1026022699000060

Cognition: Differential-geometrical view on neural networks

S. A. Buffalov

 Radio-physical Department, Tomsk State University, Lytkina 24-109, Tomsk 634034, Russia
Received 4 February 1999

Abstract

A neural network taken as a model of a trainable system appears to be nothing but a dynamical system evolving on a tangent bundle with changeable metrics. In other words to learn means to change metrics of a definite manifold.