In this paper, we consider the nonlinear dynamical behaviors of some tabuleaning neuron models. We first consider a tabu learning single neuron model.By choosing the memory decay rate as a bifurcation parameter, we prove thatHopf bifurcation occurs in the neuron. The stability of the bifurcatingperiodic solutions and the direction of the Hopf bifurcation are determined byapplying the normal form theory. We give a numerical example to verify thetheoretical analysis. Then, we demonstrate the chaotic behavior in such aneuron with sinusoidal external input, via computer simulations. Finally, westudy the chaotic behaviors in tabu learning two-neuron models, with linear andquadratic proximity functions respectively.
展开▼
