Edinburgh Research Explorer

Neural Spline Flows

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Related Edinburgh Organisations

Open Access permissions

Open

Documents

https://papers.nips.cc/paper/8969-neural-spline-flows
Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 32 (NeurIPS 2019)
PublisherCurran Associates Inc
Pages7509-7520
Number of pages12
Volume32
Publication statusPublished - 14 Dec 2019
Event33rd Conference on Neural Information Processing Systems - Vancouver Convention Centre, Vancouver, Canada
Duration: 8 Dec 201914 Dec 2019
https://neurips.cc/

Conference

Conference33rd Conference on Neural Information Processing Systems
Abbreviated titleNeurIPS 2019
CountryCanada
CityVancouver
Period8/12/1914/12/19
Internet address

Abstract

A normalizing flow models a complex probability density as an invertible transformation of a simple base density. Flows based on either coupling or autoregressive transforms both offer exact density evaluation and sampling, but rely on the parameterization of an easily invertible elementwise transformation, whose choice determines the flexibility of these models. Building upon recent work, we propose a fully-differentiable module based on monotonic rational-quadratic splines, which enhances the flexibility of both coupling and autoregressive transforms while retaining analytic invertibility. We demonstrate that neural spline flows improve density estimation, variational inference, and generative modeling of images.

Event

33rd Conference on Neural Information Processing Systems

8/12/1914/12/19

Vancouver, Canada

Event: Conference

ID: 117773185