Event Date
Speaker: Ahmed El Alaoui (Assistant Professor, Department of Statistics and Data Science, Cornell University)
Title: "Estimation, detection, and optimization from random data "
Abstract: In this talk, I will discuss some recent advances in our understanding of the fundamental limits of two ubiquitous statistical and computational tasks: 1) estimating and detecting the presence of a low rank structure buried inside a large noise matrix, and 2) optimizing a non-convex function constructed from random inputs. The first problem is a simple model for principal component analysis, and second one is at the basis of optimization-based approaches to parameter estimation. We will consider simple models: the rank-one spiked matrix model in the case of estimation, and a random quadratic function on the binary cube in the case of optimization. I will discuss the fundamental feasibility frontiers of these tasks and report on the currently best algorithmic strategies aimed at achieving them.
About the speaker: Ahmed El Alaoui is currently an Assistant Professor in department of statistics and data science at Cornell University. He received his PhD from UC Berkeley under the supervision of Michael I. Jordan and worked as a postdoctoral researcher at Stanford with Andrea Montanari. His research focuses on high-dimensional phenomena in statistics and probability theory, theoretically and computationally.
Seminar Date/Time: Thursday March 18, 4:10pm
This seminar will be delivered remotely via Zoom. To access the Zoom meeting for this seminar, please contact the instructor Xiucai Ding (xcading@ucdavis.edu) or Pete Scully (pscully@ucdavis.edu) for the meeting ID and password, stating your affiliation.