Antoine Bergel

Transforming the BCPNN Learning Rule for non-spiking units to a Learning Rule for spiking units

Abstract

In this Master Thesis Project, we aim to transform the Bayesian Confidence Propagation Neural Network (BCPNN) learning rule, including the bias term, to the domain of spiking Hodgkin-Huxley based neural networks. We also compare this learning rule to the previous BCPNN non-spiking and other spike-timing-dependent (STDP) learning rules.