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

Göltz, J.; Kriener, L.; Baumbach, A.; Billaudelle, S.; Breitwieser, O.; Cramer, B.; Dold, D.; Kungl, A. F.; Senn, W.; Schemmel, J.; Meier, K.; Petrovici, M. A. (2021). Fast and energy-efficient neuromorphic deep learning with first-spike times. Nature machine intelligence, 3(9), pp. 823-835. Springer Nature 10.1038/s42256-021-00388-x

[img]
Preview
Text
G_ltz2021Fastandene.pdf - Submitted Version
Available under License Publisher holds Copyright.
Author holds Copyright

Download (7MB) | Preview
[img] Text
s42256-021-00388-x.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.
Author holds Copyright

Download (4MB)

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 Type:

Journal Article (Original Article)

Division/Institute:

04 Faculty of Medicine > Pre-clinic Human Medicine > Institute of Physiology

Graduate School:

Graduate School for Cellular and Biomedical Sciences (GCB)

UniBE Contributor:

Göltz, Julian, Kriener, Laura Magdalena, Kungl, Akos Ferenc, Senn, Walter, Petrovici, Mihai Alexandru

Subjects:

600 Technology > 610 Medicine & health

ISSN:

2522-5839

Publisher:

Springer Nature

Language:

English

Submitter:

Stefan von Känel-Zimmermann

Date Deposited:

12 Jan 2022 12:40

Last Modified:

05 Dec 2022 15:53

Publisher DOI:

10.1038/s42256-021-00388-x

BORIS DOI:

10.48350/159882

URI:

https://boris.unibe.ch/id/eprint/159882

Actions (login required)

Edit item Edit item
Provide Feedback