Multi-Depth Temporal Fusion for Feedforward, Locally Trained Spiking Neural Networks

2026-09-29

AI Summary

This study addresses the architectural adaptation challenges of local online learning in multi-layer spiking neural networks (SNNs) by proposing a multi-depth temporal fusion framework. The core innovation lies in a depth fusion mechanism that preserves early temporal evidence, introducing it only when deep-layer features remain temporally consistent with earlier representations. By integrating residual connections and feature aggregation, we construct a convolutional backbone network based on STDP/R-STDP local learning rules and time-to-first-spike (TTFS) encoding. Experimental results demonstrate that the proposed approach significantly outperforms existing baselines on benchmarks such as CIFAR-10, simultaneously improving classification accuracy while substantially reducing event-processing overhead. The source code has been made publicly available to facilitate reproducibility.

Abstract

We propose a new spiking neural network (SNN) design to process static images and event streams using time-to-first-spike (TTFS) latencies. Our key research question is which architectural choices best accommodate local and online learning in multi-layer convolutional SNNs. This question is addressed via an original framework combining residual-like connections with multi-depth feature aggregation and consensus. The full SNN pipeline features an early-vision front end, to convert raw visual data into sparse spike latencies, a four-layer convolutional backbone trained layerwise with unsupervised spike-timing-dependent plasticity (STDP), a deterministic Multi-Depth Temporal Fusion (MDTF) and a final classifier trained with reward-modulated spike-timing-dependent plasticity (R-STDP). Rather than replacing early features in deeper layers, the proposed MDTF preserves early temporal evidence, adding sparse residual events from intermediate layers, and incorporating deeper features only when they agree in time with earlier representations. The resulting architecture is experimentally validated across MNIST, Fashion-MNIST, CIFAR-10, and N-MNIST, delivering strong classification performance under a fully local learning regime. Selective multi-depth fusion significantly outperforms traditional STDP/R-STDP baselines on higher-variability visual tasks (achieving +18.2 pp on Fashion-MNIST and +29.2 pp on CIFAR-10). Furthermore, activity-budget analyses show that the network retains high accuracy even when removing a large fraction of late or weak spike events, confirming its high data efficiency and reduced event-processing requirements. The codebase is publicly available at github.com/aidinattar/multi-depth- temporal-fusion-snn.

Problem

• Spiking Neural Networks
• Local Learning
• Online Learning
• Temporal Fusion
• STDP

Innovation

• Spiking Neural Networks
• Multi-Depth Temporal Fusion
• STDP
• Local Learning
• Time-to-First-Spike