Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Tails of Lipschitz Triangular Flows

About

We investigate the ability of popular flow based methods to capture tail-properties of a target density by studying the increasing triangular maps used in these flow methods acting on a tractable source density. We show that the density quantile functions of the source and target density provide a precise characterization of the slope of transformation required to capture tails in a target density. We further show that any Lipschitz-continuous transport map acting on a source density will result in a density with similar tail properties as the source, highlighting the trade-off between a complex source density and a sufficiently expressive transformation to capture desirable properties of a target density. Subsequently, we illustrate that flow models like Real-NVP, MAF, and Glow as implemented originally lack the ability to capture a distribution with non-Gaussian tails. We circumvent this problem by proposing tail-adaptive flows consisting of a source distribution that can be learned simultaneously with the triangular map to capture tail-properties of a target density. We perform several synthetic and real-world experiments to compliment our theoretical findings.

Priyank Jaini, Ivan Kobyzev, Yaoliang Yu, Marcus Brubaker• 2019

Related benchmarks

TaskDatasetResultRank
Heavy-tailed Flow MatchingGumbel + Gaussian copulas (test)
WP10.086
80
Flow MatchingGumbel + Gaussian alpha=1.5
Catastrophic Failure Fraction (WP1 > 1)8
20
Generative ModelingGumbel + Gaussian Median across all configurations 480 values per cell
W1^P (Pareto Margins)0.262
20
Flow MatchingGumbel + Gaussian (alpha=2.0)
Catastrophic Failure Rate17
20
Generative ModelingFama-French 5
W1 Distance0.127
5
Showing 5 of 5 rows

Other info

Follow for update