SpikeTemp: an enhanced rank-order-based learning approach for spiking neural networks with adaptive structure

Wang, J., Belatreche, A., Maguire, L.P. and McGinnity, T.M. ORCID: 0000-0002-9897-4748, 2017. SpikeTemp: an enhanced rank-order-based learning approach for spiking neural networks with adaptive structure. IEEE Transactions on Neural Networks and Learning Systems, 28 (1), pp. 30-43. ISSN 2162-237X

Preview

Text PubSub5577_McGinnity.pdf - Post-print Download (1MB) | Preview

Abstract

This paper presents an enhanced rank - order based learning algorithm, called SpikeTemp, for Spiking Neural Networks (SNNs) with a dynamically adaptive structure. The trained feed-forward SNN consists of two layers of spiking neurons: an encoding layer which temporally encodes real valued features into spatio-temporal spike patterns, and an output layer of dynamically grown neurons which perform spatio-temporal classification. Both Gaussian receptive fields and square cosine population encoding schemes are employed to encode real-valued features into spatio-temporal spike patterns. Unlike the rank-order based learning approach, SpikeTemp uses the precise times of the incoming spikes for adjusting the synaptic weights such that early spikes result in a large weight change and late spikes lead to a smaller weight change. This removes the need to rank all the incoming spikes and thus reduces the computational cost of SpikeTemp. The proposed SpikeTemp algorithm is demonstrated on several benchmark datasets and on an image recognition task. The results show that SpikeTemp can achieve better classification performance and is much faster than the existing rank-order based learning approach. In addition, the number of output neurons is much smaller when the square cosine encoding scheme is employed. Furthermore, SpikeTemp is benchmarked against a selection of existing machine learning algorithms and the results demonstrate the ability of SpikeTemp to classify different datasets after just one presentation of the training samples with comparable classification performance.

Item Type:	Journal article
Publication Title:	IEEE Transactions on Neural Networks and Learning Systems
Creators:	Wang, J., Belatreche, A., Maguire, L.P. and McGinnity, T.M.
Publisher:	IEEE
Date:	January 2017
Volume:	28
Number:	1
ISSN:	2162-237X
Identifiers:	Number Type 10.1109/TNNLS.2015.2501322 DOI
Divisions:	Schools > School of Science and Technology
Record created by:	Linda Sullivan
Date Added:	04 Jul 2016 08:47
Last Modified:	16 Oct 2017 10:56
URI:	https://irep.ntu.ac.uk/id/eprint/28063

Actions (login required)

Edit View

Views

Downloads