Dynamics of Interacting Neural Networks
Disordered Systems and Neural Networks
2007-05-23 v3
Abstract
The dynamics of interacting perceptrons is solved analytically. For a directed flow of information the system runs into a state which has a higher symmetry than the topology of the model. A symmetry breaking phase transition is found with increasing learning rate. In addition it is shown that a system of interacting perceptrons which is trained on the history of its minority decisions develops a good strategy for the problem of adaptive competition known as the Bar Problem or Minority Game.
Cite
@article{arxiv.cond-mat/9906058,
title = {Dynamics of Interacting Neural Networks},
author = {W. Kinzel and R. Metzler and I. Kanter},
journal= {arXiv preprint arXiv:cond-mat/9906058},
year = {2007}
}
Comments
9 pages, 3 figures; typos corrected, content reorganized