A Step Towards Understanding Normalizing Flows and their Likelihood Behavior

Master thesis (2024)

Authors

D.J.M. de Bruin Electrical Engineering, Mathematics and Computer Science

Contributors

Marco Loog Radboud Universiteit Nijmegen (supervisor 1)
J.C. van Gemert Pattern Recognition and Bioinformatics - EEMCS (supervisor 1)
Faculty
Electrical Engineering, Mathematics and Computer Science
Copyright: © 2024 Niels de Bruin
More Info
expand_more

Published Date

16-02-2024

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Abstract

Normalizing flows have demonstrated their ability to learn complex and high-dimensional distributions. However, the behavior of normalizing flow likelihoods are not yet fully understood, particularly when exposed to outlier data, where it has been observed that large likelihoods are often assigned to inputs that are substantially different from the training set. To better understand the likelihood behavior and outlier detection capabilities of normalizing flows, we analyze a more restricted version of the model using synthetic test data from parametric distributions, allowing access to the density of the underlying distribution.