Evasion Attacks
-
How Adversarial Examples Fool Image Classifiers
Adversarial examples follow the model's own gradient: a 0.007 per-pixel nudge made GoogLeNet call a panda a gibbon at 99.3%. How and why it works.
-
Preprocessing Defenses Fail on Depthwise-Separable Edge CNNs
A new preprint: six preprocessing defenses recover a residual CNN but not depthwise-separable ones, inverting the signal squeeze detectors read.
-
How FGSM Adversarial Attacks Work: Fast Gradient Sign Method
FGSM adds one signed gradient step to every pixel to flip a classifier's prediction. The update rule, epsilon trade-offs, and why PGD replaced it.
-
Adversarial Training Best Practices That Survive Evaluation
A practitioner's guide to adversarial training: pick a threat model, run PGD-AT correctly, catch robust overfitting, and evaluate with AutoAttack.
-
PLAA: A 92.78% NIDS Evasion Rate and Feature-Space Attacks
PLAA builds adversarial network traffic at the packet level instead of the flow level, reaching a 92.78% evasion rate against deep-learning NIDS.
-
Adversarial Examples vs. Data Poisoning
Adversarial examples attack a deployed model at inference; data poisoning attacks the model before it is deployed. Different timing, different defenses.
-
Evasion Attacks on Image Classifiers: FGSM, PGD, and C&W
A technical comparison of FGSM, PGD, and C&W explains their optimization goals, threat models, benchmark roles, limitations, and defensive implications.
-
Adversarial Transferability: How Black-Box Attacks Work
Adversarial examples transfer across models with different architectures and training sets. Why that happens, and what it means for black-box defenses.