Thermodynamic properties of theQ-state Potts-glass neural network
- 1 March 1992
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review A
- Vol. 45 (6) , 4194-4197
- https://doi.org/10.1103/physreva.45.4194
Abstract
The Q-state Potts model of neural networks, extended to include biased patterns, is studied for extensive loading α. Within the replica-symmetric approximation, mean-field equations are written down for general Q and arbitrary temperature T. The critical storage capacity is discussed for Q=3 and two classes of representative bias parameters. The complete T-α phase diagram is presented. A tricritical point is found in the spin-glass transition for Q>6, depending on α. Contrary to the Hopfield model, the critical lines do not converge to the same T as α→0. A stability analysis is made.Keywords
This publication has 21 references indexed in Scilit:
- Potts-glass model of layered feedforward neural networksPhysical Review A, 1992
- Learning grey-toned patterns in neural networksJournal of Physics A: General Physics, 1991
- The Optimal Storage Capacity for a Neural Network with Multi-State NeuronsEurophysics Letters, 1991
- Stability properties of Potts neural networks with biased patterns and low loadingJournal of Physics A: General Physics, 1991
- Storing an extensive number of grey-toned patterns in a neural network using multistate neuronsJournal of Physics A: General Physics, 1990
- The mean-field theory of a Q-state neural network modelJournal of Physics A: General Physics, 1989
- Neural networks that use three-state neuronsJournal of Physics A: General Physics, 1989
- Information processing in three-state neural networksJournal of Statistical Physics, 1989
- Discrete-state phasor neural networksPhysical Review A, 1988
- Potts-glass models of neural networksPhysical Review A, 1988