On NSL-KDD, between 19.7% and 76.2% of adversarial examples from common attacks violate the paper's network constraints, and filtering them to feasible inputs sharply lowers measured attack severity on several classifiers.
Adversarial machine learning applied to intrusion and malware scenarios: a systematic review,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Constrained Network Adversarial Attacks: Validity, Robustness, and Transferability
On NSL-KDD, between 19.7% and 76.2% of adversarial examples from common attacks violate the paper's network constraints, and filtering them to feasible inputs sharply lowers measured attack severity on several classifiers.