English

Neural network learning dynamics in a path integral framework

Statistical Mechanics 2009-11-10 v1 Disordered Systems and Neural Networks Neurons and Cognition

Abstract

A path-integral formalism is proposed for studying the dynamical evolution in time of patterns in an artificial neural network in the presence of noise. An effective cost function is constructed which determines the unique global minimum of the neural network system. The perturbative method discussed also provides a way for determining the storage capacity of the network.

Keywords

Cite

@article{arxiv.cond-mat/0308503,
  title  = {Neural network learning dynamics in a path integral framework},
  author = {J. Balakrishnan},
  journal= {arXiv preprint arXiv:cond-mat/0308503},
  year   = {2009}
}

Comments

12 pages