On Friday, April 2^nd @ 11:00AM EST, we welcome Assistant Professor of Biomedical Engineering at Johns Hopkins University, Jeremias Sulam, as he presents, "Overparameterized and Adversarially Robust Sparse Models."

ABOUT THE WEBINAR

Sparsity has been a driving force in signal & image processing and machine learning for decades. In this talk we'll explore sparse representations based on dictionary learning techniques from two perspectives: over-parameterization and adversarial robustness. First, we will characterize the surprising phenomenon that dictionary recovery can be facilitated by searching over the space of larger (over-realized/parameterized) models. This observation is general and independent of the specific dictionary learning algorithm used. We will demonstrate this observation in practice and provide a theoretical analysis of it by tying recovery measures to generalization bounds. We will further show that an efficient and provably correct distillation mechanism can be employed to recover the correct atoms from the over-realized model, consistently providing better recovery of the ground-truth model.

We will then switch gears towards the analysis of adversarial examples, focusing on the hypothesis class obtained by combining a sparsity-promoting encoder coupled with a linear classifier, and show an interesting interplay between the flexibility and stability of the (supervised) representation map and a notion of margin in the feature space. Leveraging a mild encoder gap assumption in the learned representations, we will provide a bound on the generalization error of the robust risk to L2-bounded adversarial perturbations and a robustness certificate for end-to-end classification. We will demonstrate the applicability of our analysis by computing certified accuracy on real data, and comparing with other alternatives for certified robustness. This analysis will shed light on to how to characterize this interplay for more general models.

ABOUT THE SPEAKER

Jeremias Sulam is an Assistant Professor in the Biomedical Engineering Department at Johns Hopkins University, and affiliated with the Mathematical Institute for Data Science (MINDS) and the Center for Imaging Science (CIS). He received his Bioengineering degree from UNER (Argentina) in 2013, and his PhD in Computer Science from Technion (Israel) in 2018 with Miki Elad. His research interests are focused on machine learning, signal and image processing, sparsity-inspired modeling, and their application to biomedical sciences.

ABOUT THE 2020-2021 BME WEBINAR SERIES

The Department of Biomedical Engineering at Columbia University is proud to host an annual weekly webinar series on the latest developments and research in Biomedical Engineering. The weekly series takes place on Friday mornings at 11:00AM Eastern and includes a variety of renowned academics invited from top universities to talk about their specific research and experience.