J
Jean-Michel Loubes
Researcher at Institut de Mathématiques de Toulouse
Publications - 203
Citations - 10539
Jean-Michel Loubes is an academic researcher from Institut de Mathématiques de Toulouse. The author has contributed to research in topics: Estimator & Inverse problem. The author has an hindex of 23, co-authored 184 publications receiving 9133 citations. Previous affiliations of Jean-Michel Loubes include Centre national de la recherche scientifique & Département de Mathématiques.
Papers
More filters
Journal ArticleDOI
Adaptive complexity regularization for linear inverse problems
TL;DR: In this article, a penalized method was proposed to select the optimal smoothing sequence without prior knowledge of the regularity of the function to be estimated, which is applied to Tikhonov regularization and regularization by projection.
Posted Content
Review of Mathematical frameworks for Fairness in Machine Learning
TL;DR: A review of the main fairness definitions and fair learning methodologies proposed in the literature over the last years is presented from a mathematical point of view and novel results giving the expressions of the optimal fair classifier and the optimalFair predictor in the sense of $\textit{equality of odds}$ are presented.
Journal ArticleDOI
Cochlear shape reveals that the human organ of hearing is sex-typed from birth.
José Braga,Chafik Samir,Laurent Risser,Jean Dumoncel,Didier Descouens,John Francis Thackeray,Patricia Balaresque,Anna Catherina Oettle,Jean-Michel Loubes,Anis Fradi +9 more
TL;DR: It is concluded that the human cochlea is sex-typed from an early post-natal age, and likely associated with complex evolutionary processes in modern humans for reasons not yet fully understood.
Journal ArticleDOI
A Survey of Bias in Machine Learning Through the Prism of Statistical Parity
Philippe Besse,Eustasio del Barrio,Paula Gordaliza,Paula Gordaliza,Jean-Michel Loubes,Laurent Risser +5 more
TL;DR: This article presents a mathematical framework for the fair learning problem, specifically in the binary classification setting, and proposes to quantify the presence of bias by using the standard disparate impact index on the real and well-known adult income dataset.
Journal ArticleDOI
Oracle Inequalities for a Group Lasso Procedure Applied to Generalized Linear Models in High Dimension
TL;DR: The group lasso method enables to select few groups of meaningful variables among the set of inputs and it is shown the ability of this estimator to recover good sparse approximation of the true model.