Current browse context:
cs.LG
Change to browse by:
References & Citations
Computer Science > Machine Learning
Title: Density estimation using Real NVP
(Submitted on 27 May 2016 (this version), latest version 27 Feb 2017 (v3))
Abstract: Unsupervised learning of probabilistic models is a central yet challenging problem in machine learning. Specifically, designing models with tractable learning, sampling, inference and evaluation is crucial in solving this task. We extend the space of such models using real-valued non-volume preserving (real NVP) transformations, a set of powerful invertible and learnable transformations, resulting in an unsupervised learning algorithm with exact log-likelihood computation, exact sampling, exact inference of latent variables, and an interpretable latent space. We demonstrate its ability to model natural images on four datasets through sampling, log-likelihood evaluation and latent variable manipulations.
Submission history
From: Laurent Dinh [view email][v1] Fri, 27 May 2016 21:24:32 GMT (3976kb,D)
[v2] Mon, 14 Nov 2016 21:37:10 GMT (4669kb,D)
[v3] Mon, 27 Feb 2017 23:21:10 GMT (6314kb,D)
Link back to: arXiv, form interface, contact.