Cover Image

Fast and energy-efficient neuromorphic deep learning with first-spike times

For a biological agent operating under environmental pressure, energy consumption and reaction times are of critical importance. Similarly, engineered systems are optimized for short time-to-solution and low energy-to-solution characteristics. At the level of neuronal implementation, this implies ac...

Full description

Saved in:
Bibliographic Details
Main Authors: Göltz, Julian (Author) , Kriener, Laura (Author) , Baumbach, Andreas (Author) , Billaudelle, Sebastian (Author) , Breitwieser, Oliver (Author) , Cramer, Benjamin (Author) , Dold, Dominik (Author) , Kungl, Ákos Ferenc (Author) , Senn, W. (Author) , Schemmel, Johannes (Author) , Meier, Karlheinz (Author) , Petrovici, Mihai A. (Author)
Format: Article (Journal)
Language:English
Published: 17 September 2021
In: Nature machine intelligence
Year: 2021, Volume: 3, Issue: 9, Pages: 823-835
ISSN:2522-5839
DOI:10.1038/s42256-021-00388-x
Online Access:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1038/s42256-021-00388-x
Verlag, lizenzpflichtig, Volltext: https://www.nature.com/articles/s42256-021-00388-x
Get full text
Author Notes:J. Göltz, L. Kriener, A. Baumbach, S. Billaudelle, O. Breitwieser, B. Cramer, D. Dold, A.F. Kungl, W. Senn, J. Schemmel, K. Meier and M.A. Petrovici
Description
Summary:For a biological agent operating under environmental pressure, energy consumption and reaction times are of critical importance. Similarly, engineered systems are optimized for short time-to-solution and low energy-to-solution characteristics. At the level of neuronal implementation, this implies achieving the desired results with as few and as early spikes as possible. With time-to-first-spike coding, both of these goals are inherently emerging features of learning. Here, we describe a rigorous derivation of a learning rule for such first-spike times in networks of leaky integrate-and-fire neurons, relying solely on input and output spike times, and show how this mechanism can implement error backpropagation in hierarchical spiking networks. Furthermore, we emulate our framework on the BrainScaleS-2 neuromorphic system and demonstrate its capability of harnessing the system’s speed and energy characteristics. Finally, we examine how our approach generalizes to other neuromorphic platforms by studying how its performance is affected by typical distortive effects induced by neuromorphic substrates.
Item Description:Gesehen am 20.10.2021
Physical Description:Online Resource
ISSN:2522-5839
DOI:10.1038/s42256-021-00388-x

Similar Items