There is no stopping the DEEL team from Aniti and IRT Saint Exupéry ! After a Best Paper won at the Waise 2022 conference, eight papers presented including two Best Papers at ERTS 2022, the DEEL team performs once again with eight accepted papers at the prestigious NeurIPS 2022 conference to be held on November 28th. An additional recognition that testifies to the level of expertise and quality of work of the project team.
About Neurips 2022
Founded in 1987, NeurIPS (Neural Information Processing Systems) is an international scientific conference in artificial intelligence and computational neuroscience.
This annual interdisciplinary and multidisciplinary conference includes keynote speakers, demonstrations, symposia and oral presentations, etc. It is accompanied by a professional exhibition focused on machine learning, a series of tutorials and thematic workshops that provide a less formal setting for the exchange of ideas.
Publication
Robustness
“Pay attention to your loss: understanding misconception about 1-Lipchitz neural networks”
Louis Béthune (Université Paul-Sabatier, IRIT), Thibaut Boissin (IRT Saint Exupéry), Mathieu Serrurier (Université Paul-Sabatier, IRIT), Franck Mamalet (IRT Saint Exupéry), Corentin Friedrich (IRT Saint Exupéry), Alberto González-Sanz (Université Paul-Sabatier, IMT)
Explicability
“Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure”
Paul Novello (IRT Saint Exupéry), Thomas Fel (Aniti, Université Fédérale de Toulouse), David Vigouroux (IRT Saint Exupéry)
Explicability/Neuroscience
“Aligning human and machine vision”
Thomas Fel (ANITI, Brown University), Ivan Felipe Rodriguez, Drew Linsley, Thomas Serre (Carney Institute for Brain Science, Brown University, Aniti)
“What I cannot predict, I do not understand: A human-centered Evaluation Framework for Explainability”
Julien Colin (Carney Institute for Brain Science, Brown University, Aniti), Thomas Fel (Aniti, Université Fédérale de Toulouse), Remi Cadene (Carney Institute for Brain Science, Brown University, Sorbonne Université), Thomas Serre (Carney Institute for Brain Science, Brown University, Aniti)
Reinforcement learning
“Look where you look! Saliency-guided Q-networks for visual RL tasks”
David Bertoin (IRT Saint Exupéry, INSA Toulouse), Adil Zouitine(IRT Saint Exupéry), Mehdi Zouitine (IRT Saint Exupéry), Emmanuel Rachelson (ISAE-SUPAERO, Université de Toulouse, Aniti)
Guaranties
“A general approximation lower bound in Lp norm, with applications to feed-forward neural networks”
El Mehdi Achour (IMT), Armand Foucault (IMT), Sébastien Gerchinovitz (IRT Saint Exupéry, IMT), François Malgouyres (IMT)
“Local Identifiability of Deep ReLU Neural Networks: the Theory”
Joachim Bona-Pellissier (IMT), François Malgouyres (IMT), François Bachoc (IMT)
“Automatic differentiation of nonsmooth iterative algorithms”
Jérôme Bolte (TSE), Edouard Pauwels (IRIT-ADRIA, Université Paul Sabatier), Samuel Vaiter (JAD, CNRS)
DEEL Project
The DEEL (DEpendable Explainable Learning) project involves academic and industrial partners in the development of reliable, robust, explainable and certifiable artificial intelligence technology bricks applied to critical systems.
The project covers 4 themes: reliability, explainability, certifiability and confidentiality by design.
The DEEL project is led by ANITI (Artificial and Natural Intelligence Toulouse Institute – Institut d’Intelligence Artificielle et Naturelle de Toulouse) and the IRT Saint Exupéry ; in Québec by IVADO (l’Institut de Valorisation des Données – Artificial Intelligence Research and Transfer Institute), the CRIAQ (Consortium de Recherche et d’Innovation en Aérospatiale au Québec – Consortium for Research and Innovation in Aerospace in Québec) and the IID, (Institut Intelligence et Données -Institute Intelligence and Data) at Université Laval. The DEEL project (www.deel.ai) benefits from its transatlantic collaboration France-Quebec with major industrial partners in the aeronautics, space, automotive and rail ; as well as academic partners and institutions.