Alex Damian
Kempner Research Fellow
Contact Information
Areas I Research:
About
Alex Damian is a research fellow at the Kempner Institute at Harvard University. His research focuses on the mathematical foundations of deep learning, with a focus on optimization dynamics and representation learning. He received his Ph.D. in applied and computational mathematics from Princeton University, where he was advised by Jason D. Lee, and his B.S. in mathematics from Duke University. Alex’s research has been supported by an NSF Graduate Research Fellowship and a Jane Street Graduate Research Fellowship.
Research Focus
Alex’s research is focused on the mathematical foundations of deep learning, with a focus on two interconnected areas: optimization dynamics and representation learning. His work seeks to understand how optimization algorithms, including stochastic gradient descent (SGD) and Adam, navigate the complex, high-dimensional loss landscapes that emerge during neural network training. Central to his research are three fundamental questions: What types of features can neural networks efficiently learn? How much data do they need? And how do optimization choices, including the learning rate, batch size, and momentum, shape both the training dynamics and the resulting learned representations?