Flows for simultaneous manifold learning and density estimation release_ea4bakcgrjb4tmqb257l7jiuoq

by Johann Brehmer, Kyle Cranmer

Released as a article .

2020  

Abstract

We introduce manifold-learning flows (M-flows), a new class of generative models that simultaneously learn the data manifold as well as a tractable probability density on that manifold. Combining aspects of normalizing flows, GANs, autoencoders, and energy-based models, they have the potential to represent datasets with a manifold structure more faithfully and provide handles on dimensionality reduction, denoising, and out-of-distribution detection. We argue why such models should not be trained by maximum likelihood alone and present a new training algorithm that separates manifold and density updates. In a range of experiments we demonstrate how M-flows learn the data manifold and allow for better inference than standard flows in the ambient data space.
In text/plain format

Archived Files and Locations

application/pdf  8.5 MB
file_si5ba3qofvenxjkmpinhendcx4
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2020-06-02
Version   v2
Language   en ?
arXiv  2003.13913v2
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: 926feeab-6404-4caf-9a97-8010b5aa6002
API URL: JSON