Storing information with spiking neural networks

Authors:
Radu Mirsu;Virgil Tiponut;Ioan Gavrilut
Affiliations:
Electronics Department, Politehnica University of Timisoara, Romania;Electronics Department, Politehnica University of Timisoara, Romania;Electronics Department, University of Oradea, Romania
Venue:
ICCOMP'09 Proceedings of the WSEAES 13th international conference on Computers
Year:
2009

Citing 3
Cited 0

Temporal pattern processing

The handbook of brain theory and neural networks
Spiking Neuron Models: An Introduction

Spiking Neuron Models: An Introduction
Simple model of spiking neurons

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Spiking Neural Networks encode information inside the characteristics of the spike trains they generate. There are several approaches to how this information can be coded. This paper presents coding by using spatial-temporal spike sequences. This approach is more complex compared to traditional coding schemes (rate code or phase code) but offers substantial increase in capacity due to a higher dimensionality of the representation. The paper proposes a method useful for storing spatial-temporal spike sequences with neural networks and also presents a quantitative evaluation of the network performance.