Tudor Dumitraş

Technical Report: When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attacks

2018 · arXiv (Cornell University)

Recent results suggest that attacks against supervised machine learning systems are quite effective, while defenses are easily bypassed by new attacks. However, the specifications for machine learning systems currently lack precise adversary definitions, and the …

Terminal brain damage: exposing the graceless degradation in deep neural networks under hardware fault attacks

2019 · arXiv (Cornell University)

Deep neural networks (DNNs) have been shown to tolerate brain damage: cumulative changes to the network's parameters (e.g., pruning, numerical perturbations) typically result in a graceful degradation of classification accuracy. However, the limits of this …