Voronoi Convolutional Neural Networks

10/21/2020

∙

by   Soroosh Yazdani, et al.

∙

0

∙

share

In this technical report, we investigate extending convolutional neural networks to the setting where functions are not sampled in a grid pattern. We show that by treating the samples as the average of a function within a cell, we can find a natural equivalent of most layers used in CNN. We also present an algorithm for running inference for these models exactly using standard convex geometry algorithms.