Mutual information and self-control of a fully-connected low-activity neural network

DOI10.1016/S0378-4371(00)00308-3zbMATH Open1052.68648arXivcond-mat/9912101OpenAlexW3104099958WikidataQ127330278 ScholiaQ127330278MaRDI QIDQ1581554FDOQ1581554

Authors: D. Bollé, D. D. Carreta 

Publication date: 3 October 2000

Published in: Physica A (Search for Journal in Brave)

Abstract: A self-control mechanism for the dynamics of a three-state fully-connected neural network is studied through the introduction of a time-dependent threshold. The self-adapting threshold is a function of both the neural and the pattern activity in the network. The time evolution of the order parameters is obtained on the basis of a recently developed dynamical recursive scheme. In the limit of low activity the mutual information is shown to be the relevant parameter in order to determine the retrieval quality. Due to self-control an improvement of this mutual information content as well as an increase of the storage capacity and an enlargement of the basins of attraction are found. These results are compared with numerical simulations.

Full work available at URL: https://arxiv.org/abs/cond-mat/9912101

Recommendations

• scientific article; zbMATH DE number 2095944
• Thresholds in layered neural networks with variable activity
• Dynamical properties of a randomly diluted neural network with variable activity
• Behavior of two, three and four site states self-interacting neural networks
• An optimal \(Q\)-state neural network using mutual information

zbMATH Keywords

mutual informationfully-connected neural networklow-activity neural networknetwork self-control

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• Notions of associative memory and sparse coding
• Title not available (Why is that?)
• Thresholds in layered neural networks with variable activity
• Neural Networks
• Parallel dynamics of fully connected \(Q\)-Ising neural networks
• Dynamics of temporal activity in multi-state neural networks
• Retrieval dynamics of neural networks for sparsely coded sequential patterns
• Dynamical properties of a randomly diluted neural network with variable activity
• Retrieval dynamics of associative memory of the Hopfield type

Cited In (4)

• An optimal \(Q\)-state neural network using mutual information
• Title not available (Why is that?)
• Thresholds in layered neural networks with variable activity
• Title not available (Why is that?)

This page was built for publication: Mutual information and self-control of a fully-connected low-activity neural network

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1581554)