Special Session 62: Group invariant machine learning

Max Filtering with Reflection Groups

Daniel Packer
The Ohio State University
USA
Co-Author(s):    Dustin Mixon
Abstract:
Given a finite-dimensional real inner product space $V$ and a finite subgroup $G$ of linear isometries, max filtering affords a bilipschitz Euclidean embedding of the orbit space $V/G$. We identify the max filtering maps of minimum distortion in the setting where $G$ is a reflection group. Our analysis involves an interplay between Coxeter`s classification and semidefinite programming