1 code implementation • 11 Dec 2023 • Fanny Jourdan, Louis Béthune, Agustin Picard, Laurent Risser, Nicholas Asher
In evaluation, we show that the proposed post-hoc approach significantly reduces gender-related associations in NLP models while preserving the overall performance and functionality of the models.
1 code implementation • 28 Aug 2023 • François Bachoc, Louis Béthune, Alberto González-Sanz, Jean-Michel Loubes
In this paper, we improve the learning theory of kernel distribution regression.
no code implementations • 12 Oct 2022 • François Bachoc, Louis Béthune, Alberto Gonzalez-Sanz, Jean-Michel Loubes
We present a novel kernel over the space of probability measures based on the dual formulation of optimal regularized transport.
no code implementations • NeurIPS 2023 • Mathieu Serrurier, Franck Mamalet, Thomas Fel, Louis Béthune, Thibaut Boissin
Input gradients have a pivotal role in a variety of applications, including adversarial attack algorithms for evaluating model robustness, explainable AI techniques for generating Saliency Maps, and counterfactual explanations. However, Saliency Maps generated by traditional neural networks are often noisy and provide limited insights.
no code implementations • 16 Feb 2022 • Alberto González-Sanz, Lucas de Lara, Louis Béthune, Jean-Michel Loubes
This paper introduces the first statistically consistent estimator of the optimal transport map between two probability distributions, based on neural networks.
1 code implementation • 11 Apr 2021 • Louis Béthune, Thibaut Boissin, Mathieu Serrurier, Franck Mamalet, Corentin Friedrich, Alberto González-Sanz
However they remain commonly considered as less accurate, and their properties in learning are still not fully understood.
1 code implementation • 25 Nov 2020 • Carlos Lassance, Louis Béthune, Myriam Bontonou, Mounia Hamidouche, Vincent Gripon
Measuring the generalization performance of a Deep Neural Network (DNN) without relying on a validation set is a difficult task.
1 code implementation • 8 Jul 2020 • Myriam Bontonou, Louis Béthune, Vincent Gripon
In the context of few-shot learning, one cannot measure the generalization ability of a trained classifier using validation sets, due to the small number of labeled samples.
1 code implementation • 7 Jul 2020 • Louis Béthune, Yacouba Kaloga, Pierre Borgnat, Aurélien Garivier, Amaury Habrard
We propose a novel algorithm for unsupervised graph representation learning with attributed graphs.