Optimal 1-Wasserstein distance for WGANs

DOI10.3150/23-BEJ1701zbMATH Open1545.60037MaRDI QIDQ6589580FDOQ6589580

Authors: Arthur Stephanovitch, Ugo Tanielian, Benoît Cadre, Nicolas Klutchnikoff, Gérard Biau 

Publication date: 20 August 2024

Published in: Bernoulli (Search for Journal in Brave)

Recommendations

• Some theoretical insights into Wasserstein GANs
• A geometric view of optimal transportation and generative model
• Some theoretical properties of GANs
• Multi-projection of unequal dimension optimal transport theory for generative adversary networks
• Optimal transport for generative models

zbMATH Keywords

rate of convergenceWasserstein distanceshortest pathoptimal distributionoptimal transport theoryWasserstein generative adversarial networks (WGANs)

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Optimal transportation (49Q22) Limit theorems in probability theory (60F99)

Cites Work

• Integral Probability Metrics and Their Generating Classes of Functions
• Optimal Transport
• Minkowski-type theorems and least-squares clustering
• On the rate of convergence in Wasserstein distance of the empirical measure
• Title not available (Why is that?)
• On the equality between Monge's infimum and Kantorovich's minimum in optimal mass transportation
• On the influence of the extremes of an i.i.d. sequence on the maximal spacings
• Testing the manifold hypothesis
• Title not available (Why is that?)
• Semi-discrete optimal transport: a solution procedure for the unsquared Euclidean distance case
• Title not available (Why is that?)
• Reversible Gromov-Monge sampler for simulation-based inference

Cited In (1)

• Global Sinkhorn autoencoder - optimal transport on the latent representation of the full dataset

This page was built for publication: Optimal 1-Wasserstein distance for WGANs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6589580)