Winter 2026
April 8, 2026, 4:00–5:00pm, EH4088
Speaker: March Boedihardjo (MSU)
Title: Spectral gaps and measure decompositions
Abstract: I'll introduce a new set of computable and orthogonally invariant quantities for a given probability measure on a Euclidean space. I'll show how these quantities can determine the extent to which the given probability measure can be decomposed as an equal weight mixture of two probability measures with significantly different second order statistics. Joint work with Joe Kileel and Vandy Tombs.