KIP publications

 
year 2019
author(s) Julian Göltz, Andreas Baumbach, Sebastian Billaudelle, Oliver Breitwieser, Dominik Dold, Laura Kriener, Akos Ferenc Kungl, Walter Senn, Johannes Schemmel, Karlheinz Meier, Mihai Alexandru Petrovici
title Fast and deep neuromorphic learning with time-to-first-spike coding
KIP-Nummer HD-KIP 19-117
KIP-Gruppe(n) F9
document type Paper
Keywords neuromorphic engineering, spiking neurons, physical emulation, inference, deep learning
source arXiv preprint arXiv:1912.11443
Abstract (en)

For a biological agent operating under environmental pressure, energy consumption and reaction times are of critical importance. Similarly, engineered systems also strive 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. In the time-to-first-spike coding framework, both of these goals are inherently emerging features of learning. Here, we describe a rigorous derivation of error-backpropagation-based learning for hierarchical networks of leaky integrate-and-fire neurons. We explicitly address two issues that are relevant for both biological plausibility and applicability to neuromorphic substrates by incorporating dynamics with finite time constants and by optimizing the backward pass with respect to substrate variability. This narrows the gap between previous models of first-spike-time learning and biological neuronal dynamics, thereby also enabling fast and energy-efficient inference on analog neuromorphic devices that inherit these dynamics from their biological archetypes, which we demonstrate on two generations of the BrainScaleS analog neuromorphic architecture.

bibtex
@article{goltz2019fast,
  author   = {G{\"o}ltz, Julian and Baumbach, Andreas and Billaudelle, Sebastian and Breitwieser, Oliver and Dold, Dominik and Kriener, Laura and Kungl, Akos Ferenc and Senn, Walter and Schemmel, Johannes and Meier, Karlheinz and others},
  title    = {Fast and deep neuromorphic learning with time-to-first-spike coding},
  journal  = {arXiv preprint arXiv:1912.11443},
  year     = {2019},
  volume   = {},
  pages    = {}
}
Datei pdf
URL ArXiv
up
KIP - Bibliothek
Im Neuenheimer Feld 227
Raum 3.402
69120 Heidelberg