Shahab Azarfar

Shahab Azarfar is a Postdoctoral Research Associate in the School of Data Science. Azarfar's area of research lies at the intersection of geometric deep learning and scientific machine learning. In particular, his research interests include differential geometric approach to dynamical system representation learning, equivariant neural networks, and computational optimal transport theory with application in applied physics, design of complex materials and robotics. Azarfar's educational background is in applied mathematics and mechanical engineering.

Prior to joining UVA in 2022, Azarfar worked as a Postdoctoral Fellow at the Mathematics Department, University of Toronto, and University of Western Ontario, Canada. There, he conducted research on analyzing the generative deep learning models through the lens of dynamic optimal transport theory and geometric information theory. His graduate studies were focused on modeling quantum mechanical systems using random matrix theory.

Azarfar holds a Ph.D. in Applied Mathematics and Mathematical Physics from University of Western Ontario, Canada. He also holds a M.Sc. in Applied Mathematics from Concordia University, Canada, and a M.Sc. in Mechanical Engineering from Tehran Polytechnique, Iran.