Spike-PTSD: A Bio-Plausible Adversarial Example Attack on Spiking Neural Networks via PTSD-Inspired Spike Scaling

Dimensions overall score 7.0

• Spike-PTSD achieves over 99% success rates across six datasets, three encoding types, and four models.result 95%
• Spike-PTSD systematically compromises SNN robustness.result 90%
• Spike-PTSD is a biologically inspired adversarial attack framework modeled on abnormal neural firing in Post-Traumatic Stress Disorder (PTSD).method 95%
• Spike-PTSD localizes decision-critical layers, selects neurons via hyper/hypoactivation signatures, and optimizes adversarial examples with dual objectives.method 90%
• Existing adversarial attacks often overlook SNNs' bio-plausible dynamics.limitation 85%
• Spiking Neural Networks (SNNs) are energy-efficient and biologically plausible, ideal for embedded and security-critical systems.technical 80%
• The adversarial robustness of Spiking Neural Networks remains an open research question.technical 85%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card