Scalable Learning in Structured Recurrent Spiking Neural Networks without Backpropagation

2026-05-01

AI Summary

This work addresses the challenges of scaling deep recurrent spiking neural networks (SNNs) under sparse connectivity and their reliance on backpropagation or surrogate gradients. The authors propose a purely local learning framework that eliminates the need for backpropagation by constructing a structured multilayer recurrent SNN, which integrates locally dense recurrent layers with fixed sparse small-world long-range connections. Supervised learning is achieved through a combination of population-based winner-take-all teaching signals, fixed random broadcast alignment feedback pathways, and low-dimensional modulatory neurons, all coordinated via a three-factor learning rule with eligibility traces. Experimental results demonstrate that the proposed method achieves competitive performance on standard classification benchmarks while offering notable advantages in algorithmic simplicity, computational efficiency, and hardware compatibility.

Abstract

Spiking Neural Networks (SNNs) provide a promising framework for energy-efficient and biologically grounded computation; however, scalable learning in deep recurrent architectures with sparse connectivity remains a major challenge. In this work, we propose a structured multi-layer recurrent SNN architecture composed of locally dense recurrent layers augmented with sparse small-world long-range projections to a readout population. The long-range connectivity is largely fixed, preserving routing efficiency and hardware scalability, while synaptic adaptation is performed using strictly local plasticity mechanisms. To enable supervised learning without backpropagation or surrogate gradients, we introduce a biologically motivated learning framework that combines: (i) population-based winner-take-all (WTA) teaching signals at the output layer, (ii) fixed random broadcast alignment feedback pathways, and (iii) low-dimensional modulatory neuron populations that gate synaptic updates through three-factor learning rules with eligibility traces. This design supports deep recurrent computation with sparse global communication and purely local synaptic updates. We analyze the algorithmic properties, computational complexity, and hardware feasibility of the proposed approach, and demonstrate stable learning and competitive performance on benchmark classification tasks. The results highlight the potential of structured recurrence and neuromodulatory learning to enable scalable, hardware-compatible SNN training beyond gradient-based methods.

Problem

• Spiking Neural Networks
• scalable learning
• recurrent architectures
• sparse connectivity
• backpropagation-free learning

Innovation

• Structured Recurrent SNNs
• Local Plasticity
• Three-Factor Learning
• Random Feedback Alignment
• Neuromodulatory Learning