Parallel dynamic for an extremely diluted neural network
- 21 December 1990
- journal article
- Published by IOP Publishing in Journal of Physics A: General Physics
- Vol. 23 (24) , L1323-L1337
- https://doi.org/10.1088/0305-4470/23/24/007
Abstract
The authors consider a symmetric version of the Derrida-Gardner-Zippelius model (DGZ). It is shown that in the limit of extreme dilution this modification of the DGZ model can be solved exactly. This means that for the evolution of the main overlap they obtain analytic expressions which (in contrast to the DGZ model) constitute a chain of coupled equations.Keywords
This publication has 9 references indexed in Scilit:
- Properties of sparsely connected excitatory neural networksPhysical Review A, 1990
- Dynamics of a multi-layered perceptron model : a rigorous resultJournal de Physique, 1990
- Dynamics of spin systems with randomly asymmetric bonds: Ising spins and Glauber dynamicsPhysical Review A, 1988
- Learning and forgetting on asymmetric, diluted neural networksJournal of Statistical Physics, 1987
- An Exactly Solvable Asymmetric Neural Network ModelEurophysics Letters, 1987
- On the approach of the stationary state in Kauffman's random Boolean networkJournal de Physique, 1987
- Temporal Association in Asymmetric Neural NetworksPhysical Review Letters, 1986
- The freezing transition in disordered non-equilibrium systemsJournal of Physics C: Solid State Physics, 1986
- Learning and pattern recognition in spin glass modelsZeitschrift für Physik B Condensed Matter, 1985