MorphSNN: Adaptive Graph Diffusion and Structural Plasticity for Spiking Neural Networks

2026-03-15

AI Summary

Existing spiking neural networks (SNNs) struggle to adapt to complex environments due to their static architectures, which lack brain-inspired dynamic topology and lateral interactions. To address this limitation, this work proposes MorphSNN, a novel framework that, for the first time, integrates non-synaptic graph diffusion mechanisms with spatiotemporal structural plasticity (STSP) into SNNs. This integration enables undirected signal propagation and instance-driven self-evolving network reconfiguration. The resulting dynamic topology not only enhances adaptability but also allows the evolved network structure to serve as an intrinsic distributional fingerprint for unsupervised out-of-distribution (OOD) detection without auxiliary training. Evaluated on datasets such as N-Caltech101, MorphSNN achieves 83.35% accuracy within only five time steps, outperforming current state-of-the-art methods.

Technology Category

Cognitive Modeling & Cognitive Systems: Neural Spike CodingMachine Learning: Evolutionary LearningComputer Vision: Diffusion Models for Vision

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSocial Networks and Social Media: Influence propagation, information diffusion, and the prediction on networksWeb Mining and Content Analysis: Models for Web evolution

Abstract

Spiking Neural Networks (SNNs) currently face a critical bottleneck: while individual neurons exhibit dynamic biological properties, their macro-scopic architectures remain confined within conventional connectivity patterns that are static and hierarchical. This discrepancy between neuron-level dynamics and network-level fixed connectivity eliminates critical brain-like lateral interactions, limiting adaptability in changing environments. To address this, we propose MorphSNN, a backbone framework inspired by biological non-synaptic diffusion and structural plasticity. Specifically, we introduce a Graph Diffusion (GD)mechanism to facilitate efficient undirected signal propagation, complementing the feedforward hierarchy. Furthermore, it incorporates a Spatio-Temporal Structural Plasticity (STSP) mechanism, endowing the network with the capability for instance-specific, dynamic topological reorganization, thereby overcoming the limitations of fixed topologies. Experiments demonstrate that MorphSNN achieves state-of-the-art accuracy on static and neuromorphic datasets; for instance, it reaches 83.35% accuracy on N-Caltech101 with only 5 timesteps. More importantly, its self-evolving topology functions as an intrinsic distribution fingerprint, enabling superior Out-of- Distribution (OOD) detection without auxiliary training. The code is available at anonymous.4open.science/r/MorphSNN-B0BC.

Problem

• Spiking Neural Networks
• structural plasticity
• graph diffusion
• fixed topology
• adaptability

Innovation

• Graph Diffusion
• Structural Plasticity
• Spiking Neural Networks
• Dynamic Topology
• Out-of-Distribution Detection