Search

Recently, multi-stable Neural Networks (NN) with exponential number of attractors have been presented and analyzed theoretically; however, the learning process of the parameters of these systems while considering stability conditions and specifications of real world problems has not been studied. In this paper, a new class of multi-stable NNs using sinusoidal dynamics with exponential number of attractors is introduced. The sufficient conditions for multi-stability of the proposed system are posed using Lyapunov theorem. In comparison to the other methods in this class of multi-stable NNs, the proposed method is used as a classifier by applying a learning process with respect to the topological information of data and conditions of Lyapunov multi-stability. The proposed NN is applied on both synthetic and real world datasets with an accuracy comparable to classical classifiers., (Copyright © 2015 Elsevier Ltd. All rights reserved.)

Some of neuropathologies are believed to be related to abnormal synchronization of neurons. In the line of therapy, designing effective deep brain stimulators to suppress the pathological synchrony among neuronal ensembles is a challenge of high clinical relevance. The stimulation should be able to disrupt the synchrony in the presence of latencies due to imperfect knowledge about parameters of a neuronal ensemble and stimulation impacts on the ensemble. We propose an adaptive desynchronizing deep brain stimulator capable of dealing with these uncertainties. We analyze the collective behavior of the stimulated neuronal ensemble and show that, using the designed stimulator, the resulting asynchronous state is stable. Simulation results reveal the efficiency of the proposed technique., (Copyright © 2014 Elsevier Ltd. All rights reserved.)