Analysis of SpikeProp Convergence with Alternative Spike Response Functions

Vaenthan Thiruvarudchelvan, James Crane, Terence Bossomaier

Research output: Book chapter/Published conference paperConference paperpeer-review

4 Citations (Scopus)
13 Downloads (Pure)

Abstract

SpikeProp is a supervised learning algorithm for spiking neural networks analogous to backpropagation. Like backpropagation, it may fail to converge for particular networks, parameters and datasets. However there are several behaviours and additional failure modes unique to SpikeProp which have notbeen explicitly outlined in the literature. These factors hinder the adoption of SpikeProp for general machine learning use. In this paper we examine the mathematics of SpikeProp in detail and identify the various causes of failure therein. The analysis implies that applying certain constraints on parameterslike initial weights can improve the rates of convergence. It also suggests that alternative spike response functions could improve the learning rate and reduce the number of convergence failures. We tested two alternative functions and found these predictions to be true.
Original languageEnglish
Title of host publicationProceedings of the 2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages98-105
Number of pages8
ISBN (Electronic)9781467359016
DOIs
Publication statusPublished - 2013
Event2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI) - Grand Copthorne Waterfront Hotel, Singapore, Singapore
Duration: 16 Apr 201319 Apr 2013

Conference

Conference2013 IEEE Symposium on Foundations of Computational Intelligence (FOCI)
CountrySingapore
CitySingapore
Period16/04/1319/04/13

Fingerprint Dive into the research topics of 'Analysis of SpikeProp Convergence with Alternative Spike Response Functions'. Together they form a unique fingerprint.

Cite this