AIMS Home

Group invariant machine learning

Organizer(s):
Name:	Affiliation:	Country:
Jameson Cahill	University of North Carolina Wilmington	USA
Dustin Mixon	Ohio State University	USA

Abstract:

Many data sets display symmetry which can be modeled as a group acting on the data. Accounting for this symmetry can lead to improved performance for a variety of tasks. Such problems require insight from a various areas of mathematics.

List of approved abstract
Name	Affiliation	Title
Radu Balan	University of Maryland	Coorbit Representations of Homogeneous Metric Spaces
Ben Blum-Smith	Johns Hopkins University	Invariant machine learning on point clouds
Dan Edidin	University of Missouri	Separating orbits using invariants of low degree
Joseph W Iverson	Iowa State University	Numerically stable group invariants
Oscar Mickelin	Princeton University	Fast Principal Component Analysis for Cryo-EM Images
Daniel Packer	The Ohio State University	Max Filtering with Reflection Groups
Michael Perlmutter	UCLA	Geometric Scattering on Measure Spaces
Yousef Qaddura	The Ohio State University	Injectivity, stability, and positive definiteness of max filtering
Nathaniel Strawn	Georgetown University	Universality of Andrews Networks