CogniSNN: Enabling Neuron-Expandability, Pathway-Reusability, and Dynamic-Configurability with Random Graph Architectures in Spiking Neural Networks

2025-12-12

AI Summary

Conventional spiking neural networks (SNNs) inherit rigid hierarchical architectures from artificial neural networks (ANNs), overlooking the scalability, pathway reuse, and dynamic reconfigurability conferred by the brain’s stochastic neuronal connectivity. Method: This paper proposes a brain-inspired SNN paradigm centered on a Random Graph Architecture (RGA), replacing fixed layers with a flexible graph-based topology that supports dynamic neuron addition/removal, adaptive critical pathway reuse, and real-time network restructuring. It introduces three novel components: a pure-spike residual mechanism, KP-LwF—a knowledge-preserving lifelong learning algorithm—and Dynamic Growth Learning (DGL), a strategy for structured network expansion. Contribution/Results: The framework achieves state-of-the-art accuracy on neuromorphic benchmarks and Tiny-ImageNet, demonstrates significantly enhanced continual learning capability, improved robustness, and effectively alleviates fixed-timing constraints inherent in neuromorphic hardware.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Spiking neural networks (SNNs), regarded as the third generation of artificial neural networks, are expected to bridge the gap between artificial intelligence and computational neuroscience. However, most mainstream SNN research directly adopts the rigid, chain-like hierarchical architecture of traditional artificial neural networks (ANNs), ignoring key structural characteristics of the brain. Biological neurons are stochastically interconnected, forming complex neural pathways that exhibit Neuron-Expandability, Pathway-Reusability, and Dynamic-Configurability. In this paper, we introduce a new SNN paradigm, named Cognition-aware SNN (CogniSNN), by incorporating Random Graph Architecture (RGA). Furthermore, we address the issues of network degradation and dimensional mismatch in deep pathways by introducing an improved pure spiking residual mechanism alongside an adaptive pooling strategy. Then, we design a Key Pathway-based Learning without Forgetting (KP-LwF) approach, which selectively reuses critical neural pathways while retaining historical knowledge, enabling efficient multi-task transfer. Finally, we propose a Dynamic Growth Learning (DGL) algorithm that allows neurons and synapses to grow dynamically along the internal temporal dimension. Extensive experiments demonstrate that CogniSNN achieves performance comparable to, or even surpassing, current state-of-the-art SNNs on neuromorphic datasets and Tiny-ImageNet. The Pathway-Reusability enhances the network's continuous learning capability across different scenarios, while the dynamic growth algorithm improves robustness against interference and mitigates the fixed-timestep constraints during neuromorphic chip deployment. This work demonstrates the potential of SNNs with random graph structures in advancing brain-inspired intelligence and lays the foundation for their practical application on neuromorphic hardware.

Problem

• Introduces a random graph architecture for brain-like structural features in SNNs
• Addresses network degradation and mismatch in deep pathways via improved mechanisms
• Enables dynamic growth and selective reuse for efficient multi-task learning

Innovation

• Introduces Random Graph Architecture for brain-like connectivity
• Uses Key Pathway Learning to reuse critical neural pathways
• Implements Dynamic Growth Learning for adaptive neuron expansion