BioLeaF: A Bio-plausible Learning Framework for Training of Spiking Neural Networks

TL;DR

This paper introduces a biologically plausible learning framework for spiking neural networks that achieves accuracy comparable to backpropagation-based methods by using a new microcircuit architecture and local learning rules.

Contribution

It proposes a novel bio-plausible architecture and learning rules for training multi-layer SNNs, bridging the gap with BP-based methods in accuracy.

Findings

Achieves comparable accuracy to BP-based training methods.

Introduces a microcircuit architecture with local feedback connections.

Provides an optimization interpretation linking it to backpropagation.

Abstract

Our brain consists of biological neurons encoding information through accurate spike timing, yet both the architecture and learning rules of our brain remain largely unknown. Comparing to the recent development of backpropagation-based (BP-based) methods that are able to train spiking neural networks (SNNs) with high accuracy, biologically plausible methods are still in their infancy. In this work, we wish to answer the question of whether it is possible to attain comparable accuracy of SNNs trained by BP-based rules with bio-plausible mechanisms. We propose a new bio-plausible learning framework, consisting of two components: a new architecture, and its supporting learning rules. With two types of cells and four types of synaptic connections, the proposed local microcircuit architecture can compute and propagate error signals through local feedback connections and support training of multi-layers SNNs with a globally defined spiking error function. Under our microcircuit architecture, we employ the Spike-Timing-Dependent-Plasticity (STDP) rule operating in local compartments to update synaptic weights and achieve supervised learning in a biologically plausible manner. Finally, We interpret the proposed framework from an optimization point of view and show the equivalence between it and the BP-based rules under a special circumstance. Our experiments show that the proposed framework demonstrates learning accuracy comparable to BP-based rules and may provide new insights on how learning is orchestrated in biological systems.