The wasserstein proximal gradient algorithm

Adil Salim, Anna Korba, Giulia Luise

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

Abstract

Wasserstein gradient flows are continuous time dynamics that define curves of steepest descent to minimize an objective function over the space of probability measures (i.e., the Wasserstein space). This objective is typically a divergence w.r.t. a fixed target distribution. In recent years, these continuous time dynamics have been used to study the convergence of machine learning algorithms aiming at approximating a probability distribution. However, the discrete-time behavior of these algorithms might differ from the continuous time dynamics. Besides, although discretized gradient flows have been proposed in the literature, little is known about their minimization power. In this work, we propose a Forward Backward (FB) discretization scheme that can tackle the case where the objective function is the sum of a smooth and a nonsmooth geodesically convex terms. Using techniques from convex optimization and optimal transport, we analyze the FB scheme as a minimization algorithm on the Wasserstein space. More precisely, we show under mild assumptions that the FB scheme has convergence guarantees similar to the proximal gradient algorithm in Euclidean spaces.
Original languageEnglish (US)
Title of host publication34th Conference on Neural Information Processing Systems, NeurIPS 2020
PublisherNeural information processing systems foundation
StatePublished - Jan 1 2020

Fingerprint

Dive into the research topics of 'The wasserstein proximal gradient algorithm'. Together they form a unique fingerprint.

Cite this