Brain-Inspired Learning on Neuromorphic Substrates

TL;DR

This paper develops a mathematical framework connecting online gradient-based learning algorithms with biologically plausible rules for training spiking neural networks on neuromorphic hardware, addressing key challenges in real-world applications.

Contribution

It introduces a novel framework linking RTRL and biological learning rules, with a sparse approximation to enhance efficiency and applicability on neuromorphic substrates.

Findings

Sparse block-diagonal Jacobian approximation improves computational efficiency.

Empirical results show good learning performance with the proposed method.

Framework bridges gap between synaptic plasticity and deep learning algorithms.

Abstract

Neuromorphic hardware strives to emulate brain-like neural networks and thus holds the promise for scalable, low-power information processing on temporal data streams. Yet, to solve real-world problems, these networks need to be trained. However, training on neuromorphic substrates creates significant challenges due to the offline character and the required non-local computations of gradient-based learning algorithms. This article provides a mathematical framework for the design of practical online learning algorithms for neuromorphic substrates. Specifically, we show a direct connection between Real-Time Recurrent Learning (RTRL), an online algorithm for computing gradients in conventional Recurrent Neural Networks (RNNs), and biologically plausible learning rules for training Spiking Neural Networks (SNNs). Further, we motivate a sparse approximation based on block-diagonal Jacobians, which reduces the algorithm's computational complexity, diminishes the non-local information requirements, and empirically leads to good learning performance, thereby improving its applicability to neuromorphic substrates. In summary, our framework bridges the gap between synaptic plasticity and gradient-based approaches from deep learning and lays the foundations for powerful information processing on future neuromorphic hardware systems.