Join us for the next speaker in our ECE graduate seminar series, Akram Awad, who will present "Robust Approach for Transfer Learning via Optimal Transport."
ABSTRACT: Distributionally robust optimization (DRO) mitigates the effect of distributional uncertainty in supervised learning by optimizing over an uncertainty ball of distributions, typically centered around the empirical distribution of the training sample. In this talk, DRO is considered for the problem of unsupervised domain adaptation (UDA). In classical UDA, the goal is to adapt a model trained on a labeled source domain to a new, unlabeled target domain. Modifying classical DRO to UDA settings by enlarging the uncertainty radius around the source to include the target can lead to excessive regularization. To mitigate this, Akram Awad proposes utilizing optimal transport to transport the source domain to a vicinity of the target and construct the DRO problem around the transported samples, thereby ensuring a small uncertainty radius in DRO with high likelihood of including the true target. His numerical experiments validate the superiority of his method over existing robust approaches.
BIO: Akram Awad is pursuing a doctorate in electrical engineering at UCF. He received his bachelor’s degree in electrical engineering from Jordan University of Science and Technology, Irbid, Jordan, in 2016.and his master’s degree in electrical and computer engineering from Oakland University, Rochester, Michigan in 2020. His research interests include robust learning, domain adaptation and reinforcement learning. He is a student member of IEEE.
Read More