Research indexBook
Adversarial Attack Detection using Normalizing Flows
Gasimov
KITopen26 Aug 2026Open access
NobleIDark:/48914/w1/20261105/F6755823DOI10.5445/ir/1000196498
Adversarial Attack Detection using Normalizing Flows
Gasimov · KITopen · 2026
NobleIDark:/48914/w1/20261105/F6755823DOI10.5445/ir/1000196498
Free to read at the publisher
You can read the abstract here, or the full text at the publisher.
Abstract
The deployment of Deep Neural Networks within the perception pipelines of autonomous driving systems is critically hindered by their vulnerability to visually inconspicuous adversarial pertur- bations. In safety-critical applications such as Traffic Sign Recognition, these mathematical ma- nipulations can force highly confident misclassifications, bypassing traditional Out-of-Distribution detection mechanisms that rely on compromised softmax probability distributions. To secure these perception systems without degrading their baseline accuracy, this thesis proposes and engineers an unsupervised, parallel runtime safety monitor based on exact density estimation utilizing the multi-scale Glow Normalizing Flow architecture. The central scientific contribution of this research is a comprehensive, mathematically grounded ablation study investigating the optimal topological input space for generative anomaly detectors. The study directly compares an Input-Only flow architecture, which models the exact log-likelihood of the raw pixel space, against a novel Joint Input-Feature architecture. The joint configuration ex- tracts intermediate semantic representations from the deep hidden layers of a frozen target classifier, spatially aligns them, and models the massively high-dimensional concatenated tensor, theoretically attempting to fuse low-level statistical anomalies with high-level semantic inconsistencies. Extensive empirical evaluations were conducted across multiple deep learning backbones and highly diverse traffic sign datasets against optimal white-box L∞ attacks. The results conclusively demonstrate that the Input-Only exact density estimator successfully identifies the high-frequency structural disruptions induced by digital adversarial perturbations, achieving robust detection mar- gins. Conversely, the Joint Input-Feature architecture exhibits a severe degradation in detection efficacy. This counter-intuitive failure is formally attributed to the Semantic Alignment Paradox, wherein successfully optimized adversarial features perfectly mimic the natural semantic profile of the target class, thereby tricking the generative model into assigning a high likelihood score. Furthermore, the computational overhead analysis proves that the massive quadratic inflation of Floating Point Operations and inference latency required to process the concatenated semantic fea- tures strictly prohibits real-time deployment. Ultimately, this thesis mathematically and empirically establishes that the lightweight Input-Only topology is the strictly superior, computationally viable paradigm for defending latency-constrained autonomous systems against digital adversarial threats.