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

TL;DR

This paper introduces a supervised learning method for spiking neural networks that enhances robustness to device mismatch and noise, enabling efficient deployment on mixed-signal neuromorphic hardware without per-device calibration.

Contribution

The authors propose a novel supervised training approach for SNNs that mimics a pre-trained dynamical system, improving robustness to hardware mismatch and noise.

Findings

Our method outperforms common training alternatives in robustness.

The approach enables deployment without per-device calibration.

Demonstrated on memory-dependent classification tasks.

Abstract

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 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.