Double spike temporal coding for Spiking Neural Networks
Keywords
Loading...
Authors
Issue Date
2022-06-20
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A lot of research has been done in the field of Artificial Neural Networks, but research is just starting to explore the possibilities of Spiking Neural Networks. This work expands on the paper “Biologically plausible Learning of text Representation with Spiking Neural Networks” by exploring the influence of double spike versus single spike temporal coding on learning. Double spike temporal coding is represented by this paper’s proposed synaptic spike trace that acts as a weight updating inhibitor. The models were trained according to a winner-takes-all strategy on the 20-newsgroup dataset. Both model’s encoder neuron dynamics are based on the Leaky Integrate-and-Fire principle, with synaptic plasticity defined by the Hebbian Learning rule of Spike Timing Dependent Plasticity. Comparing the accuracy of both models, shows that double spike temporal, as implemented by this paper, does not have a positive effect on the accuracy of text classification for Spiking Neural Networks.
Description
Citation
Supervisor
Faculty
Faculteit der Sociale Wetenschappen