Matthias Löwe et Franck Vermet.


Titre: "The storage capacity of the Blume-Emery-Griffiths neural network."


Résumé : We analyze the so called Blume-Emery-Griffiths (BEG) neural network at zero temperature. We give an upper bound on its storage capacity provided that we want the stored patterns to be fixed points of the retrieval dynamics. We also discuss a more liberal notion of storage capacity introduced by Newman [1] in the context of the Hopfield model [2]. We show that, similar to the findings in the neural networks literature, the BEG model with this notion of storage capacity can store a number of patterns proportional to the number of neurons in the model.

[1] C. Newman (1988). Memory capacity in neural networks; Neural Networks 1, 223-238.

[2] J.J. Hopfield (1982). Neural networks and physical systems with emergent collective computational abilities; Proc. nat. Acad. Sci. USA 79, 2554-2558.