Supervised training of spiking neural networks for robust deployment on mixed-signal neuromorphic processors

Mixed-signal analog/digital circuits emulate spiking neurons and synapses with extremely high energy efficiency, an approach known as “neuromorphic engineering”. However, analog circuits are sensitive to process-induced variation among transistors in a chip (“device mismatch”). For neuromorphic impl...

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports 2021-12, Vol.11 (1), p.23376-23376, Article 23376
Main Authors: Büchel, Julian, Zendrikov, Dmitrii, Solinas, Sergio, Indiveri, Giacomo, Muir, Dylan R.
Format: Article
Language:English
Subjects:
639/166/987
639/705/117
Action Potentials
Algorithms
Animals
Biomimetics - instrumentation
Calibration
Control theory
Energy efficiency
Firing pattern
Humanities and Social Sciences
Humans
Learning
Mimicry
Models, Neurological
multidisciplinary
Neural networks
Neural Networks, Computer
Neurons - physiology
Science
Science (multidisciplinary)
Supervised Machine Learning
Synapses
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Mixed-signal analog/digital circuits emulate spiking neurons and synapses with extremely high energy efficiency, an approach known as “neuromorphic engineering”. However, analog circuits are sensitive to process-induced variation among transistors in a chip (“device mismatch”). For neuromorphic implementation of Spiking Neural Networks (SNNs), mismatch causes parameter variation between identically-configured neurons and synapses. Each chip exhibits a different distribution of neural parameters, causing deployed networks to respond differently between chips. Current solutions to mitigate mismatch based on per-chip calibration or on-chip learning entail increased design complexity, area and cost, making deployment of neuromorphic devices expensive and difficult. Here we present a supervised learning approach that produces SNNs with high robustness to mismatch and other common sources of noise. Our method trains SNNs to perform temporal classification tasks by mimicking a pre-trained dynamical system, using a local learning rule from non-linear control theory. We demonstrate our method on two tasks requiring temporal memory, and measure the robustness of our approach to several forms of noise and mismatch. We show that our approach is more robust than common alternatives for training SNNs. Our method provides robust deployment of pre-trained networks on mixed-signal neuromorphic hardware, without requiring per-device training or calibration.
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-021-02779-x