Spatio-Temporal Pattern Selectivity from Homeostatic Hebbian Plasticity
Klaus Pawelzik,Mohammad Dehghani Habibabadi · COSYNE 2022
It is an open question to what extent neural coding and computation are based on precise patterns of spikes. Theoretically individual neurons can serve as detectors for given spatio-temporal spike patterns, however, this requires supervised adjustment of their input synapses. It is not known if existing activity dependent synaptic plasticity mechanisms can lead to unsupervised emergence of spatio-temporal pattern selectivity. Here, a combination of realistic mechanisms is demonstrated to self-organize the synaptic input e?cacies such that neurons become detectors of patterns repeating in the i
Thu, Mar 17, 2022