66 citations · 362 across the 77 of their papers we have counts for
46 papers · 1 filter
Wasserstein Spatial Depth
François Bachoc, Alberto González-Sanz, Jean-Michel Loubes +1
Modeling observations as random distributions embedded within Wasserstein spaces is becoming increasingly popular across scientific fields, as it captures the variability and geome…
Improved learning theory for kernel distribution regression with two-stage sampling
François Bachoc, Louis Béthune, Alberto González-Sanz +1
The distribution regression problem encompasses many important statistics and machine learning tasks, and arises in a large range of applications. Among various existing approaches…
On the coalitional decomposition of parameters of interest
Marouane Il Idrissi, Nicolas Bousquet, Fabrice Gamboa +2
Understanding the behavior of a black-box model with probabilistic inputs can be based on the decomposition of a parameter of interest (e.g., its variance) into contributions attri…
Diffeomorphic Registration using Sinkhorn Divergences
Lucas de Lara, Alberto González-Sanz, Jean-Michel Loubes
The diffeomorphic registration framework enables to define an optimal matching function between two probability measures with respect to a data-fidelity loss function. The non conv…
An improved central limit theorem and fast convergence rates for entropic transportation costs
Eustasio del Barrio, Alberto Gonzalez-Sanz, Jean-Michel Loubes +1
We prove a central limit theorem for the entropic transportation cost between subgaussian probability measures, centered at the population cost. This is the first result which allo…
Fairness seen as Global Sensitivity Analysis
Clément Bénesse, Fabrice Gamboa, Jean-Michel Loubes +1
Ensuring that a predictor is not biased against a sensible feature is the key of Fairness learning. Conversely, Global Sensitivity Analysis is used in numerous contexts to monitor…
The statistical effect of entropic regularization in optimal transportation
Eustasio del Barrio, Jean-Michel Loubes
We propose to tackle the problem of understanding the effect of regularization in Sinkhorn algotihms. In the case of Gaussian distributions we provide a closed form for the regular…
Minimax optimal goodness-of-fit testing for densities and multinomials under a local differential privacy constraint
Joseph Lam-Weil, Béatrice Laurent, Jean-Michel Loubes
Finding anonymization mechanisms to protect personal data is at the heart of recent machine learning research. Here, we consider the consequences of local differential privacy cons…
A Central Limit Theorem for transportation cost with applications to Fairness Assessment in Machine Learning
Eustasio del Barrio, Paula Gordaliza, Jean-Michel Loubes
We provide a Central Limit Theorem for the Monge-Kantorovich distance between two empirical distributions with size and , for for observations on the re…
Obtaining fairness using optimal transport theory
Eustasio del Barrio, Fabrice Gamboa, Paula Gordaliza +1
Statistical algorithms are usually helping in making decisions in many aspects of our lives. But, how do we know if these algorithms are biased and commit unfair discrimination of…
Risk Measures and Credit Risk Under the Beta-Kotz Distribution
Jean-Michel Loubes, M Andrea Arias-Serna, Francisco Caro-Lopera
This paper considers the use for Value-at-Risk computations of the so-called Beta-Kotz distribution based on a general family of distributions including the classical Gaussian mode…
Distribution regression model with a Reproducing Kernel Hilbert Space approach
Thi Thien Trang Bui, J-M Loubes, Laurent Risser +1
In this paper, we introduce a new distribution regression model for probability distributions. This model is based on a Reproducing Kernel Hilbert Space (RKHS) regression framework…
A data driven trimming procedure for robust classification
Marina Antolín, Eustasio Del Barrio, Jean-Michel Loubes
Classification rules can be severely affected by the presence of disturbing observations in the training sample. Looking for an optimal classifier with such data may lead to unnece…
Central limit theorem and bootstrap procedure for Wasserstein's variations with an application to structural relationships between distributions
Eustasio Del Barrio, Paula Gordaliza, Hélène Lescornel +1
Wasserstein barycenters and variance-like criteria based on the Wasserstein distance are used in many problems to analyze the homogeneity of collections of distributions and struct…
A new gene co-expression network analysis based on Core Structure Detection (CSD)
A-C Brunet, J-M Azais, J-M Loubes +2
We propose a novel method to cluster gene networks. Based on a dissimilarity built using correlation structures, we consider networks that connect all the genes based on the streng…
Maximum likelihood estimation for a bivariate Gaussian process under fixed domain asymptotics
Daira Velandia, François Bachoc, Moreno Bevilacqua +2
We consider maximum likelihood estimation with data from a bivariate Gaussian process with a separable exponential covariance model under fixed domain asymptotic. We first characte…
A statistical analysis of a deformation model with Wasserstein barycenters : estimation procedure and goodness of fit test
Eustasio Del Barrio, Hélène Lescornel, Jean-Michel Loubes
We propose a study of a distribution registration model for general deformation functions. In this framework, we provide estimators of the deformations as well as a goodness of fit…
Existence and Consistency of Wasserstein Barycenters
Thibaut Le Gouic, Jean-Michel Loubes
In this paper, based on the Fr{é}chet mean, we define a notion of barycenter corresponding to a usual notion of statistical mean. We prove the existence of Wasserstein barycenters…
A unified framework for the study of the PLS estimator's properties
Mélanie Blazère, Fabrice Gamboa, Jean-Michel Loubes
In this paper we propose a new approach to study the properties of the Partial Least Squares (PLS) estimator. This approach relies on the link between PLS and discrete orthogonal p…
A Kriging procedure for processes indexed by graphs
Thibault Espinasse, Jean-Michel Loubes
We provide a new kriging procedure of processes on graphs. Based on the construction of Gaussian random processes indexed by graphs, we extend to this framework the usual linear pr…
PLS: a new statistical insight through the prism of orthogonal polynomials
Mélanie Blazère, Fabrice Gamboa, Jean-Michel Loubes
Partial Least Square (PLS) is a dimension reduction method used to remove multicollinearities in a regression model. However contrary to Principal Components Analysis (PCA) the PLS…
Big Data Analytics - Retour vers le Futur 3; De Statisticien à Data Scientist
Philippe Besse, Aurélien Garivier, Jean-Michel Loubes
The rapid evolution of information systems managing more and more voluminous data has caused profound paradigm shifts in the job of statistician, becoming successively data miner,…
Group Lasso for generalized linear models in high dimension
Mélanie Blazère, Jean-Michel Loubes, Fabrice Gamboa
Nowadays an increasing amount of data is available and we have to deal with models in high dimension (number of covariates much larger than the sample size). Under sparsity assumpt…
A robust algorithm for template curve estimation based on manifold embedding
Chloé Dimeglio, Santiago Gallón, Jean-Michel Loubes +1
This paper considers the problem of finding a meaningful template function that represents the common pattern of a sample of curves. To address this issue, a novel algorithm based…