Statistical guarantees for generative models without domination

MaRDI QIDQ7028606zbMATH OpenFDO

Authors Nicolas Schreuder, Victor-Emmanuel Brunel, Arnak S. Dalalyan

Publication date 11 February 2025

Full work available at URL https://proceedings.mlr.press/v132/schreuder21a.html

zbMATH Keywords

generative modelrisk boundsmoothness class

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Artificial neural networks and deep learning (68T07)

Cites work

• Bounding the expectation of the supremum of empirical processes indexed by Hölder classes
• Confidence regions and minimax rates in outlier-robust estimation on the probability simplex
• Convergence and concentration of empirical measures under Wasserstein distance in unbounded functional spaces
• Formulae for high derivatives of composite functions
• Integral Probability Metrics and Their Generating Classes of Functions
• On the empirical estimation of integral probability metrics
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Some theoretical properties of GANs
• The Speed of Mean Glivenko-Cantelli Convergence

Cited in

(6)

• Statistical guarantees for distribution estimation of contaminated data via DNN-based MoM-GANs
• Deep generative models: complexity, dimensionality, and approximation
• On the statistical properties of generative adversarial models for low intrinsic data dimension
• Rates of convergence for nonparametric estimation of singular distributions using generative adversarial networks
• A supervised deep learning method for nonparametric density estimation
• Distribution estimation via flow matching with Lipschitz guarantees

This page was built for publication: Statistical guarantees for generative models without domination

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7028606)