Image evolution in Potts-glass neural networks
D. Bollé, F. Mallezie · Journal of Physics A Mathematical and General · 1989
Neural networks of the Potts type, where neurons can occupy q>or=2 discrete states, are considered for couplings which need not be symmetric. The time evolution of the macroscopic overlap between the instantaneous microscopic state of the system and the learned patterns is studied for small values of q and for a finite (small) number of patterns. Retrieval and limit-cycle behaviour in this type of model is discussed in some detail.