The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks

Zenke F, Vogels TP. 2021. The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks. Neural Computation. 33(4), 899–925.

Download
OA 2021_NeuralComputation_Zenke.pdf 1.61 MB [Published Version]

Journal Article | Published | English

Scopus indexed
Author
Zenke, Friedemann ; Vogels, Tim PISTA

Corresponding author has ISTA affiliation

Department
Abstract
Brains process information in spiking neural networks. Their intricate connections shape the diverse functions these networks perform. In comparison, the functional capabilities of models of spiking networks are still rudimentary. This shortcoming is mainly due to the lack of insight and practical algorithms to construct the necessary connectivity. Any such algorithm typically attempts to build networks by iteratively reducing the error compared to a desired output. But assigning credit to hidden units in multi-layered spiking networks has remained challenging due to the non-differentiable nonlinearity of spikes. To avoid this issue, one can employ surrogate gradients to discover the required connectivity in spiking network models. However, the choice of a surrogate is not unique, raising the question of how its implementation influences the effectiveness of the method. Here, we use numerical simulations to systematically study how essential design parameters of surrogate gradients impact learning performance on a range of classification problems. We show that surrogate gradient learning is robust to different shapes of underlying surrogate derivatives, but the choice of the derivative’s scale can substantially affect learning performance. When we combine surrogate gradients with a suitable activity regularization technique, robust information processing can be achieved in spiking networks even at the sparse activity limit. Our study provides a systematic account of the remarkable robustness of surrogate gradient learning and serves as a practical guide to model functional spiking neural networks.
Publishing Year
Date Published
2021-03-01
Journal Title
Neural Computation
Publisher
MIT Press
Acknowledgement
F.Z. was supported by the Wellcome Trust (110124/Z/15/Z) and the Novartis Research Foundation. T.P.V. was supported by a Wellcome Trust Sir Henry Dale Research fellowship (WT100000), a Wellcome Trust Senior Research Fellowship (214316/Z/18/Z), and an ERC Consolidator Grant SYNAPSEEK.
Volume
33
Issue
4
Page
899-925
ISSN
eISSN
IST-REx-ID

Cite this

Zenke F, Vogels TP. The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks. Neural Computation. 2021;33(4):899-925. doi:10.1162/neco_a_01367
Zenke, F., & Vogels, T. P. (2021). The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks. Neural Computation. MIT Press. https://doi.org/10.1162/neco_a_01367
Zenke, Friedemann, and Tim P Vogels. “The Remarkable Robustness of Surrogate Gradient Learning for Instilling Complex Function in Spiking Neural Networks.” Neural Computation. MIT Press, 2021. https://doi.org/10.1162/neco_a_01367.
F. Zenke and T. P. Vogels, “The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks,” Neural Computation, vol. 33, no. 4. MIT Press, pp. 899–925, 2021.
Zenke F, Vogels TP. 2021. The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks. Neural Computation. 33(4), 899–925.
Zenke, Friedemann, and Tim P. Vogels. “The Remarkable Robustness of Surrogate Gradient Learning for Instilling Complex Function in Spiking Neural Networks.” Neural Computation, vol. 33, no. 4, MIT Press, 2021, pp. 899–925, doi:10.1162/neco_a_01367.
All files available under the following license(s):
Creative Commons Attribution 4.0 International Public License (CC-BY 4.0):
Main File(s)
Access Level
OA Open Access
Date Uploaded
2022-04-08
MD5 Checksum
eac5a51c24c8989ae7cf9ae32ec3bc95


Export

Marked Publications

Open Data ISTA Research Explorer

Web of Science

View record in Web of Science®

Sources

PMID: 33513328
PubMed | Europe PMC

Search this title in

Google Scholar