Iterative Hard Thresholding for Sparse Generalized Linear Models

Abstract: The first step to understand non-linear models in machine learning is the study of generalized linear models that include linear regression, logistic regression, half-space learning, and two-layer neural networks as special cases. In this talk we discuss some algorithmic approaches for parameter learning in sparse generalized linear models (GLM). In particular, we focus on an iterative hard thresholding algorithm for sparse GLMs. We present convergence results for the binary iterative hard thresholding for the 1-bit compressed sensing problem, a very successful high-dimensional signal acquisition and recovery technique.

 

Speaker Bio: Arya Mazumdar is an Associate Professor of Data Science and Computer Science in UC San Diego. He is the Deputy Director and the Associate Director for Research in the NSF AI Institute TILOS, and also the UCSD Site-Lead of NSF TRIPODS Institute EnCORE. Arya obtained his Ph.D. degree from University of Maryland, College Park specializing in information theory. Subsequently Arya was a postdoctoral scholar at Massachusetts Institute of Technology, an assistant professor in University of Minnesota, and an assistant followed by associate professor in University of Massachusetts Amherst.  Arya is a recipient of a Distinguished Dissertation Award for his Ph.D. thesis, the NSF CAREER award, an EURASIP Best Paper Award, and the ISIT Jack K. Wolf Student Paper Award. He is also a Distinguished Lecturer of the IEEE Information Theory Society, 2023-24. He is currently serving as an Associate Editor for the IEEE Transactions on Information Theory and as an Area editor for Now Publishers Foundation and Trends in Communication and Information Theory. Arya’s research interests include  information theory, coding theory, statistical learning and optimization.