NSPDI-SNN: An efficient lightweight SNN based on nonlinear synaptic pruning and dendritic integration

2025-08-29

AI Summary

Existing spiking neural networks (SNNs) oversimplify dendritic structures, failing to capture their nonlinear integration and high sparsity—limiting spatiotemporal representation capability and computational efficiency. To address this, we propose NSPDI-SNN, the first SNN framework that jointly integrates nonlinear dendritic integration (NDI) with heterogeneous synaptic pruning (NSP), incorporating a dynamic dendritic spine pruning mechanism featuring variable-state transitions. This design enables efficient structural sparsification while preserving architectural lightness. Our approach significantly enhances information transmission efficiency and topological flexibility: even at >90% sparsity, accuracy degradation remains below 0.5%. NSPDI-SNN achieves state-of-the-art performance across diverse benchmarks—including DVS128 Gesture, CIFAR10-DVS, speech recognition, and reinforcement learning-based maze navigation—demonstrating the efficacy of biologically inspired sparse computation paradigms.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Deep Neural Architectures and Foundation Models
• Natural Language Processing: Learning & Optimization for NLP

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: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Spiking neural networks (SNNs) are artificial neural networks based on simulated biological neurons and have attracted much attention in recent artificial intelligence technology studies. The dendrites in biological neurons have efficient information processing ability and computational power; however, the neurons of SNNs rarely match the complex structure of the dendrites. Inspired by the nonlinear structure and highly sparse properties of neuronal dendrites, in this study, we propose an efficient, lightweight SNN method with nonlinear pruning and dendritic integration (NSPDI-SNN). In this method, we introduce nonlinear dendritic integration (NDI) to improve the representation of the spatiotemporal information of neurons. We implement heterogeneous state transition ratios of dendritic spines and construct a new and flexible nonlinear synaptic pruning (NSP) method to achieve the high sparsity of SNN. We conducted systematic experiments on three benchmark datasets (DVS128 Gesture, CIFAR10-DVS, and CIFAR10) and extended the evaluation to two complex tasks (speech recognition and reinforcement learning-based maze navigation task). Across all tasks, NSPDI-SNN consistently achieved high sparsity with minimal performance degradation. In particular, our method achieved the best experimental results on all three event stream datasets. Further analysis showed that NSPDI significantly improved the efficiency of synaptic information transfer as sparsity increased. In conclusion, our results indicate that the complex structure and nonlinear computation of neuronal dendrites provide a promising approach for developing efficient SNN methods.

Problem

• Improving spiking neural network efficiency with dendritic structure modeling
• Achieving high sparsity in SNNs through nonlinear synaptic pruning
• Enhancing spatiotemporal information representation with dendritic integration

Innovation

• Nonlinear dendritic integration for spatiotemporal representation
• Heterogeneous state transition ratios in dendritic spines
• Flexible nonlinear synaptic pruning achieving high sparsity