Topology-aware design of spiking neural networks via modular graph architectures

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: 6-10× reduction in training epochs required

• Primary Outcome: Classification accuracy and training efficiency on benchmark datasets

• Nnt Or Sensitivity: 99.57% accuracy on MNIST, >92% on CIFAR-10

• Confidence Interval: Not reported

Clinical Application

High feasibility for implementation in research and development environments with appropriate computational resources Relevant to Australian research institutions and technology companies developing AI hardware and neuromorphic systems, particularly for applications in remote sensing and edge computing where energy efficiency is critical Researchers and engineers developing neuromorphic computing systems, edge AI applications, and energy-efficient neural networks

Abstract

Spiking Neural Networks (SNNs) offer a biologically plausible and energy-efficient alternative to traditional artificial neural networks (ANNs), yet their design remains constrained by limited architectural flexibility and slow training dynamics. In this work, we introduce a novel SNN framework that leverages modular graph-based topologies and explicit synaptic delays to significantly enhance both training efficiency and classification performance. Our architecture, TANet-Tiny, incorporates structured graph stages with up to 32 nodes and diverse community-driven connectivity patterns derived from KMeans clustering, Louvain modularity, and Watts-Strogatz small-world models. We integrate these topologies into a topology-aware search space and explore them via a Spatio-Temporal Topology Sampling (STTS) approach, enabling the discovery of high-performing networks without exhaustive search. Experimental results on MNIST, CIFAR-10, and CIFAR-100 demonstrate that our modular designs achieve state-of-the-art accuracy while requiring 6-10 × fewer training epochs, with top-1 accuracy reaching 99.57% on MNIST and over 92% on CIFAR-10, all with reduced parameter counts. We introduce an accuracy-per-epoch metric to quantify training efficiency and show that modularity, rather than network size, is the critical driver of performance. This work lays the groundwork for scalable, interpretable, and low-latency SNN architectures suitable for deployment in neuromorphic and edge computing environments.

References

1. 1.Motaghian, F., Nazari, S., Dominguez-Morales, J. P., & Jafari, R. (2026). Topology-aware design of spiking neural networks via modular graph architectures. PLoS One. https://doi.org/10.1371/journal.pone.0344997