Electronic realizations of neurons are of great interest as building blocks for neuromorphic computation. Electronic neurons should send signals into the input and output lines when subject to an input signal exceeding a given threshold, in such a way that they may affect all other parts of a neural network. Here, we propose a design for a neuron that is based on molecular-electronics components and thus promises a very high level of integration. We employ the Monte Carlo technique to simulate typical time evolutions of this system and thereby show that it indeed functions as a neuron.
C Timm and M Di Ventra
Click for full article
C Timm and M Di Ventra
Click for full article
No comments:
Post a Comment