Normalizing Flows on Tori and Spheres
Machine Learning
2020-07-02 v2 Machine Learning
Abstract
Normalizing flows are a powerful tool for building expressive distributions in high dimensions. So far, most of the literature has concentrated on learning flows on Euclidean spaces. Some problems however, such as those involving angles, are defined on spaces with more complex geometries, such as tori or spheres. In this paper, we propose and compare expressive and numerically stable flows on such spaces. Our flows are built recursively on the dimension of the space, starting from flows on circles, closed intervals or spheres.
Cite
@article{arxiv.2002.02428,
title = {Normalizing Flows on Tori and Spheres},
author = {Danilo Jimenez Rezende and George Papamakarios and Sébastien Racanière and Michael S. Albergo and Gurtej Kanwar and Phiala E. Shanahan and Kyle Cranmer},
journal= {arXiv preprint arXiv:2002.02428},
year = {2020}
}
Comments
Accepted to the International Conference on Machine Learning (ICML) 2020