Random Graphs and Nonlinear Spectral Gaps
Offered By: Hausdorff Center for Mathematics via YouTube
Course Description
Overview
Explore a lecture on random graphs and nonlinear spectral gaps presented by Pandelis Dodos at the Hausdorff Center for Mathematics. Delve into the challenging problem posed by Pisier and Mendel-Naor regarding regular expander graphs and discrete Poincaré inequalities for functions with values in Banach spaces. Examine the positive results for Banach spaces with unconditional bases and cotype q ≥ 2, and understand the transfer argument by Naor/Ozawa and nonlinear embedding by Odell-Schlumprecht. Learn about the concept of long-range expansion in regular graphs and its implications. Discover two key findings: the high probability of uniformly random d-regular graphs satisfying long-range expansion, and the discrete Poincaré inequality for functions in Banach spaces with unconditional bases and cotype q in graphs with long-range expansion. Gain insights into the nearly optimal Poincaré constant estimate proportional to q¹⁰ in this joint work with Dylan Altschuler, Konstantin Tikhomirov, and Konstantinos Tyros.
Syllabus
Pandelis Dodos: Random graphs and nonlinear spectral gaps
Taught by
Hausdorff Center for Mathematics
Related Courses
Теория вероятностей для начинающихMoscow Institute of Physics and Technology via Coursera Universality of Random Graphs - Gal Kronenberg
Institute for Advanced Study via YouTube The Science of Networks - C4 Public Lectures
Santa Fe Institute via YouTube Optimization of the Sherrington-Kirkpatrick Hamiltonian
IEEE via YouTube Fast Uniform Generation of Random Graphs with Given Degree Sequences
IEEE via YouTube