Spike train propagation in Hybrid Artificial Neural Network (HANN)

TL;DR

This paper explores how a Hybrid Artificial Neural Network (HANN) can generate spike trains, aiming to make artificial neural network simulations more biologically realistic by incorporating physical phenomena like intermittency.

Contribution

It demonstrates the potential of HANN to produce biologically plausible spike trains based on physical dynamical concepts rather than traditional mathematical algorithms.

Findings

HANN can generate spike trains resembling biological neurons.

Intermittency phenomena are key to biological neural dynamics.

HANN operates based on physical notions rather than purely mathematical models.

Abstract

The spikes train is an important step in order to the artificial neural network (ANN) give us simulations more close to the reality i.e the operation of the biological neural network. Based on in previous our work that the HANN can to produce critical and tricritical intermittencies we investigate in present work the possibility of the Spike train production from the HANN. So the operation of ANN does not would based in mathematical algorithm of machine learning but the operation of a ANN could be based in physical notions as the phenomenon of intermittency. As we have shown the real biological neurons is a Dynamical system which present the intermittent dynamic type I.