attack arXiv Apr 2, 2026 · 4d ago
Lingxin Jin, Wei Jiang, Maregu Assefa Habtie et al. · University of Electronic Science and Technology · Khalifa University
Bio-inspired adversarial attack on Spiking Neural Networks achieving 99% success by exploiting PTSD-like abnormal neuron firing patterns
Input Manipulation Attack vision
Spiking Neural Networks (SNNs) are energy-efficient and biologically plausible, ideal for embedded and security-critical systems, yet their adversarial robustness remains open. Existing adversarial attacks often overlook SNNs' bio-plausible dynamics. We propose Spike-PTSD, a biologically inspired adversarial attack framework modeled on abnormal neural firing in Post-Traumatic Stress Disorder (PTSD). It localizes decision-critical layers, selects neurons via hyper/hypoactivation signatures, and optimizes adversarial examples with dual objectives. Across six datasets, three encoding types, and four models, Spike-PTSD achieves over 99% success rates, systematically compromising SNN robustness. Code: https://github.com/bluefier/Spike-PTSD.
traditional_ml University of Electronic Science and Technology · Khalifa University
ML Security Papers