Marissa Connor
Software Engineering Institute
Marissa Connor is a machine learning research scientist in the Adversarial Machine Learning Lab within the AI Division. She has broad experience working on adversarial machine learning topics including data poisoning and evasion attacks against image classification (supervised and semi-supervised learning), object detection, and large language models.
Before joining the SEI at the end of 2023, Connor worked for Embedded Intelligence, a small machine learning start up. As a principal investigator for the team working on DARPA’s Guaranteeing AI Robustness Against Deception (GARD) program, she worked on developing quality assurance tools for identifying poisoned images. Connor earned her PhD in electrical engineering from Georgia Tech where her research focused on learning generative representations of natural data transformations.
