An optimal Q-state neural network using mutual information

DOI10.1016/S0375-9601(02)00437-1zbMATH Open0995.82045arXivcond-mat/0112093OpenAlexW2013614098MaRDI QIDQ1598596FDOQ1598596

Authors: D. Bollé, T. Verbeiren 

Publication date: 23 May 2002

Published in: Physics Letters. A (Search for Journal in Brave)

Abstract: Starting from the mutual information we present a method in order to find a hamiltonian for a fully connected neural network model with an arbitrary, finite number of neuron states, Q. For small initial correlations between the neurons and the patterns it leads to optimal retrieval performance. For binary neurons, Q=2, and biased patterns we recover the Hopfield model. For three-state neurons, Q=3, we find back the recently introduced Blume-Emery-Griffiths network hamiltonian. We derive its phase diagram and compare it with those of related three-state models. We find that the retrieval region is the largest.

Full work available at URL: https://arxiv.org/abs/cond-mat/0112093

Recommendations

• Mutual information and self-control of a fully-connected low-activity neural network
• Thermodynamic properties of fully connected Q-Ising neural networks
• scientific article; zbMATH DE number 1928604
• Behavior of two, three and four site states self-interacting neural networks
• The capacity of \(q\)-state Potts neural networks with parallel retrieval dynamics

zbMATH Keywords

phase diagramHamiltonianoptimal retrieval performance

Mathematics Subject Classification ID

Neural nets applied to problems in time-dependent statistical mechanics (82C32)

Cites Work

• Neural networks and physical systems with emergent collective computational abilities
• Mutual information and self-control of a fully-connected low-activity neural network
• Thermodynamic properties of fully connected Q-Ising neural networks

Cited In (2)

• Parallel dynamics of the fully connected Blume-Emery-Griffiths neural network
• Mutual information and self-control of a fully-connected low-activity neural network

This page was built for publication: An optimal \(Q\)-state neural network using mutual information

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1598596)