Dynamics of interacting neural networks

DOI10.1088/0305-4470/33/14/101MaRDI QIDQ4507647zbMATH OpenOpenAlexFDO

Authors Ido Kanter, Wolfgang Kinzel, Richard Metzler

Publication date 8 October 2000

Published in Journal of Physics A: Mathematical and General (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/cond-mat/9906058

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Software, source code, etc. for problems pertaining to statistical mechanics (82-04) Neural nets applied to problems in time-dependent statistical mechanics (82C32) Neural networks for/in biological studies, artificial life and related topics (92B20)

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.

Recommendations

• Statistical physics of interacting neural networks
• The connections of large perceptrons
• Statistical mechanics of training in neural networks
• Phase transitions and learning in neural networks
• The space of interactions in neural network models

Cited in

(10)

• Evolutionary minority games: the benefits of imitation
• Estimating the interaction graph of stochastic neural dynamics
• Moral foundations in an interacting neural networks society: a statistical mechanics analysis
• Emergent dynamics in a network of recurrently coupled neurons
• Collective dynamics of neural network with distance dependent field coupling
• Neural cryptography with queries
• Out-of-equilibrium dynamical mean-field equations for the perceptron model
• Statistical physics of interacting neural networks
• Interacting dreaming neural networks
• Reactive means in the iterated Prisoner's dilemma

This page was built for publication: Dynamics of interacting neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4507647)