Poisoning Network Flow Classifiers
This work addresses security vulnerabilities in ML-based network monitoring systems, offering incremental improvements in attack stealth and effectiveness for cybersecurity applications.
The paper tackles the problem of poisoning attacks on network traffic flow classifiers by developing a clean-label backdoor attack method that uses model interpretability to craft effective triggers at low poisoning rates and generative models for stealth, achieving feasibility in scenarios like malicious communication detection.
As machine learning (ML) classifiers increasingly oversee the automated monitoring of network traffic, studying their resilience against adversarial attacks becomes critical. This paper focuses on poisoning attacks, specifically backdoor attacks, against network traffic flow classifiers. We investigate the challenging scenario of clean-label poisoning where the adversary's capabilities are constrained to tampering only with the training data - without the ability to arbitrarily modify the training labels or any other component of the training process. We describe a trigger crafting strategy that leverages model interpretability techniques to generate trigger patterns that are effective even at very low poisoning rates. Finally, we design novel strategies to generate stealthy triggers, including an approach based on generative Bayesian network models, with the goal of minimizing the conspicuousness of the trigger, and thus making detection of an ongoing poisoning campaign more challenging. Our findings provide significant insights into the feasibility of poisoning attacks on network traffic classifiers used in multiple scenarios, including detecting malicious communication and application classification.