. Expected Exploitability: Predicting the Development of Functional Vulnerability Exploits. Preprint, 2021.

Preprint PDF

. Exploring Adversarial Examples in Malware Detection. S&P’19 Deep Learning and Security Workshop, 2019.

Preprint PDF Slides Venue

. Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks. NeurIPS’18, 2018.

Preprint PDF Code

. When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attacks. USENIX’18, 2018.

PDF Code Slides Technical Report Presentation

. Too Big to FAIL: What You Need to Know Before Attacking a Machine Learning System. SPW’18, 2018.


. Summoning Demons: The Pursuit of Exploitable Bugs in Machine Learning. NeurIPS’16 Workshop on Reliable Machine Learning in the Wild, 2016.

Preprint PDF Slides Presentation Venue Media Coverage


  • osuciu AT umd DOT edu
  • 5112 Brendan Iribe Center, College Park, MD 20740