Ph.D. Candidate: Nathaniel Thomas
Research Advisor: Patrick Hayden
Date: Wednesday, June 19th, 2019
Location: Gates 176 (B wing)
Title: Euclidean-equivariant functions on three-dimensional point clouds
I present a type of neural network that is locally equivariant to 3D rotations, translations, and permutations of points at each layer. Local 3D rotation equivariance removes the need for data augmentation to identify features in arbitrary orientations. These networks use convolution filters built from spherical harmonics; due to the mathematical consequences of this filter choice, each layer accepts as input (and guarantees as output) scalars, vectors, and higher-order tensors, in the geometric sense of these terms. I show applications of these networks to supervised learning tasks in structural biology.