Neuromorphic Cybersecurity with Semi-supervised Lifelong Learning

2025-08-06

AI Summary

To address catastrophic forgetting, low energy efficiency, and insufficient biological interpretability in lifelong-learning network intrusion detection systems (NIDS), this paper proposes a brain-inspired collaborative spiking neural network (SNN) architecture. The architecture integrates static and dynamic SNN modules, incorporates a Grow-When-Required (GWR) mechanism to enable structural plasticity-driven model expansion, and introduces an adaptive spike-timing-dependent plasticity (STDP) rule to support unsupervised incremental learning. Sparse spiking simulations are conducted on the Intel Lava framework, achieving 85.3% overall accuracy on the UNSW-NB15 dataset. The approach significantly mitigates catastrophic forgetting while maintaining high adaptability and ultra-low power consumption. This work establishes a novel paradigm for continual security monitoring on neuromorphic hardware, bridging biologically plausible learning mechanisms with practical cybersecurity requirements.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Life-Long and Continual Learning
• Natural Language Processing: Safety and Robustness

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Inspired by the brain's hierarchical processing and energy efficiency, this paper presents a Spiking Neural Network (SNN) architecture for lifelong Network Intrusion Detection System (NIDS). The proposed system first employs an efficient static SNN to identify potential intrusions, which then activates an adaptive dynamic SNN responsible for classifying the specific attack type. Mimicking biological adaptation, the dynamic classifier utilizes Grow When Required (GWR)-inspired structural plasticity and a novel Adaptive Spike-Timing-Dependent Plasticity (Ad-STDP) learning rule. These bio-plausible mechanisms enable the network to learn new threats incrementally while preserving existing knowledge. Tested on the UNSW-NB15 benchmark in a continual learning setting, the architecture demonstrates robust adaptation, reduced catastrophic forgetting, and achieves $85.3$% overall accuracy. Furthermore, simulations using the Intel Lava framework confirm high operational sparsity, highlighting the potential for low-power deployment on neuromorphic hardware.

Problem

• Develops SNN for lifelong Network Intrusion Detection System (NIDS).
• Enables incremental learning of new threats without forgetting.
• Achieves high accuracy and low-power neuromorphic deployment.

Innovation

• Spiking Neural Network for lifelong NIDS
• GWR-inspired structural plasticity adaptation
• Ad-STDP learning rule for incremental threats