Marwa El Halabi

I am currently a visiting researcher at Mila - Quebec AI Institute and a CIFAR AI Safety postdoctoral fellow, working with Gauthier Gidel. Previously, I was a research scientist at the Samsung AI Lab Montreal, an academic-style research lab situated within Mila. Before that, I was a postdoctoral researcher in the Machine Learning Group  at MIT working with Stefanie Jegelka. 

I completed my PhD in 2018 at the Computer and Communication Sciences department at EPFL where I was advised by Volkan Cevher.  I received my B.Eng. in Computer and Communications Engineering in 2012 from the American University of Beirut.

My research interests include discrete and continuous optimization in machine learning, AI safety, and AI for social good applications. My current work focuses on LLM safety and (non-)submodular optimization. In the past, I have worked on (non-)submodular optimization, (non-)convex optimization, neural network compression, and structured sparsity.