"Deep Generative models and Inverse Problems", by Alex Dimakis
Deep Generative models and Inverse Problems, 16 June,听18:00 through MS Teams听, by 听Professor of听Electrical and Computer Engineering Department,听University of Texas at Austin, USA
Modern deep generative models like GANs, VAEs and invertible flows are demonstrating听excellent performance in representing high-dimensional distributions, especially for images.听We will show how they can be used to solve inverse problems like denoising, filling missing听data, and medical imaging. We generalize compressed sensing theory beyond sparsity, extending听Restricted Isometries to sets created by deep generative models. We will present the general听framework, recent results and open problems in this space.
听
Biography
Alex Dimakis is a Professor at the Electrical and Computer Engineering department,听University of Texas at Austin and the co-director of the NSF National institute on the听Foundations of Machine Learning. He received his Ph.D. from UC Berkeley and the听Diploma degree from the National Technical University of Athens. He received听several awards including the James Massey Award, NSF Career, a Google research听award, the Eli Jury dissertation award and the joint Information Theory and Communications听Society Best Paper Award. His research interests include information theory, coding听theory and machine learning.
