Jeremy Cohen

Jeremy Cohen

I am currently working in the LLM industry. Previously, I was a Research Fellow at the Flatiron Institute, in the Center for Computational Mathematics. Before that, I did a PhD in the machine learning department at Carnegie Mellon, co-advised by Zico Kolter and Ameet Talwalkar.

My research has been aimed at connecting theory and practice in deep learning. The ultimate goal is to turn deep learning from a kind of alchemy into a mature engineering discipline.

In particular, I've studied the dynamics of optimization in deep learning. Neural networks are trained using optimization algorithms, and understanding the behavior of these algorithms could help us understand and improve the training process.

Featured Publications