Recursive Construction of Periodoc Steady State for Neural Networks
Martı́n Matamala · HAL (Le Centre pour la Communication Scientifique Directe) · 1993
We present a strategy in order to build neural networks with long steady state periodic behavior. This strategy allows us to obtain 2^n non equivalent neural networks of size n, when the equivalence relation is the dynamical systems one. As a particular case, we build a neural networks with n neurons admitting a cycle of period 2^n.