Abstract
We study the multivariate deconvolution problem of recovering the distribution of a signal from independent and identically distributed observations additively contaminated with random errors having known distribution. For errors with ordinary smooth distribution, we recast the multidimensional problem as a one-dimensional problem leveraging the equivalence between the -Wasserstein and the max-sliced -Wasserstein metrics and derive an inversion inequality relating the -Wasserstein distance between two distributions of the signal to the -distance between the corresponding mixture densities of the observations. This smoothing inequality outperforms existing inversion inequalities. We apply it to derive -Wasserstein rates of convergence for the distribution of the signal. As an application to the Bayesian framework, we consider -Wasserstein deconvolution with the Laplace noise in dimension one using a Dirichlet process mixture of normal densities as a prior measure for the mixing distribution. We construct an adaptive approximation of the sampling density by convolving the Laplace density with a well-chosen mixture of Gaussian densities and show that the posterior measure contracts at a nearly minimax-optimal rate, up to a log-factor, in the -distance. The rate automatically adapts to the unknown Sobolev regularity of the mixing density, thus leading to a new Bayesian adaptive estimation procedure over the full scale of regularity levels. We illustrate the utility of the inversion inequality also in a frequentist setting by showing that a minimum distance estimator attains the minimax convergence rates for -Wasserstein deconvolution in any dimension , lower bounds being derived here.
Funding Statement
The authors gratefully acknowledge financial support from the Institut Henri Poincaré (IHP), Sorbonne Université (Paris), within the RIP program “Bayesian Wasserstein deconvolution” that took place in 2019 at the IHP-Centre Émile Borel.
Catia Scricciolo has also been partially supported by Università di Verona and MUR—Prin 2022—Grant No. 2022CLTYP4, funded by the European Union—Next Generation EU.
The project leading to this work has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 834175).
Acknowledgements
The authors would like to thank the Editor, two Associate Editors and three anonymous referees for constructive and valuable comments that helped improve the original manuscript.
Judith Rousseau is also affiliated to the Ceremade, CMRS, UMR 7534, Université Paris-Dauphine, PSL University.
Catia Scricciolo is a member of the “Gruppo Nazionale per l’Analisi Matematica, la Probabilità e le loro Applicazioni (GNAMPA)” of the Istituto Nazionale di Alta Matematica (INdAM). She wishes to dedicate this work to her mother and sister, Emilia, with deep love and immense gratitude.
Citation
Judith Rousseau. Catia Scricciolo. "Wasserstein convergence in Bayesian and frequentist deconvolution models." Ann. Statist. 52 (4) 1691 - 1715, August 2024. https://doi.org/10.1214/24-AOS2413
Information