Publications

Computer Science

Deep Anti-Regularized Ensembles provide reliable out-of-distribution uncertainty quantification

We consider the problem of uncertainty quantification in high dimensional regression and classification for which deep ensemble have proven to be promising methods. Recent observations have shown that deep ensemble often return overconfident…

Antoine de Mathelin, Francois Deheeger, Mathilde Mougeot, Nicolas Vayatis

Computer Science

ADAPT : Awesome Domain Adaptation Python Toolbox

In this paper, we introduce the ADAPT library, an open source Python API providing the implementation of the main transfer learning and domain adaptation methods. The library is designed with a user friendly approach to facilitate the access to…

Antoine de Mathelin, Mounir Atiq, Guillaume Richard, Alejandro de la Concha, Mouad Yachouti, François Deheeger, Mathilde Mougeot, Nicolas Vayatis

Computer Science

Maximum Weight Entropy

This paper deals with uncertainty quantification and out-of-distribution detection in deep learning using Bayesian and ensemble methods. It proposes a practical solution to the lack of prediction diversity observed recently for standard approaches…

Antoine de Mathelin, François Deheeger, Mathilde Mougeot, Nicolas Vayatis

Life Sciences

Synergy of Mutation-Induced Effects in Human Vitamin K Epoxide Reductase: Perspectives and Challenges for Allo-Network Modulator Design

The human Vitamin K Epoxide Reductase Complex (hVKORC1), a key enzyme transforming vitamin K into the form necessary for blood clotting, requires for its activation the reducing equivalents delivered by its redox partner through thiol-disulfide…

Marina Botnari, Luba Tchertanov

International Journal of Molecular Sciences - 2024

Life Sciences

Influence of Psycho-emotional Factors on Motor Control : Cerebral Mechanism and Behavioral Response Underlying (Motiv)Action

The objective of this research topic is to regroup a collection of manuscripts investigating the relationship between the psycho-emotional components and motor control using multidisciplinary approaches. Movement performance can be influenced by…

Thierry Lelard, Pierre-Paul Vidal, Guillaume Leonard

Mathematics

Quasi-local and frequency robust preconditioners for the Helmholtz first-kind integral equations on the disk

We propose preconditioners for the Helmholtz scattering problems by a planar, disk-shaped screen in $\R^3$. Those preconditioners are approximations of the square-roots of some partial differential operators acting on the screen. Their matrix-vector…

François Alouges, Martin Averseng

ESAIM: Mathematical Modelling and Numerical Analysis - 2024

Other Statistics

Stein Boltzmann Sampling: A Variational Approach for Global Optimization

In this paper, we present a flow-based method for global optimization of continuous Sobolev functions, called Stein Boltzmann Sampling (SBS). SBS initializes uniformly a number of particles representing candidate solutions, then uses the Stein…

Gaëtan Serré, Argyris Kalogeratos, Nicolas Vayatis

Humanities and Social Sciences

Que la transition écologique soit, et la transition numérique fût.

Marie Garin

Statistics

L’épuration professionnelle du monde du spectacle à la Libération. Histoire et sciences mathématiques

Karine Le Bail, Julien Randon-Furling

Annales. Histoire, Sciences sociales - 2024

Machine Learning

Conformal Approach To Gaussian Process Surrogate Evaluation With Coverage Guarantees

Gaussian processes (GPs) are a Bayesian machine learning approach widely used to construct surrogate models for the uncertainty quantification of computer simulation codes in industrial applications. It provides both a mean predictor and an estimate…

Edgar Jaber, Vincent Blot, Nicolas Brunel, Vincent Chabridon, Emmanuel Remy, Bertrand Iooss, Didier Lucor, Mathilde Mougeot, Alessandro Leite