1.1k citations
- Sorbonne UniversitéFR42 papers
- Université Pierre-et-Marie-CurieFR28 papers
- Université Paris CitéFR17 papers
- Laboratoire de Probabilités et Modèles AléatoiresFR14 papers
- Centre d'Économie de la SorbonneFR13 papers
- Centre National de la Recherche ScientifiqueFR10 papers
- Maison des Sciences sociales et des Humanités Ange GuépinFR9 papers
- Laboratoire de Mathématiques Blaise PascalFR8 papers
- Université Paris 1 Panthéon-SorbonneFR8 papers
- École Normale Supérieure - PSLFR7 papers
- Mathématiques Appliquées à Paris 5FR7 papers
- Psychiatric Medicine AssociatesUS7 papers
171 papers
On the contraction properties of Sinkhorn semigroups
O. Deniz Akyildiz, Pierre del Moral, Joaquin Miguez
We develop a novel stability theory for Sinkhorn semigroups based on Lyapunov techniques and quantitative contraction coefficients, and establish exponential convergence of Sinkhor…
New probabilistic methods for physics
Guy Cirier
We present how a probabilistic model can describe the asymptotic behavior of the iterations, with applications for ODE and approach of some problems in mechanics in .
A probabilistic approach of the Poincar{é}-Bendixon's problem in
Guy Cirier
We present how a probabilistic model can describe the asymptotic behaviour of the iterations, especially for ODE with an approach of the Poincaré-Bendixon's problem in $\mathbb{R}^…
Interpretable Random Forests via Rule Extraction
Clément Bénard, Gérard Biau, Sébastien da Veiga +1
We introduce SIRUS (Stable and Interpretable RUle Set) for regression, a stable rule learning algorithm which takes the form of a short and simple list of rules. State-of-the-art l…
On choking and ingestion hazards for children in the United States
Frédéric Chabolle, Paul Deheuvels
The risks of Unintentional Suffocation injuries in the U.S. must be reconsidered in view of the existent mortality and morbidity statistics. In particular, fatal injuries due to th…
A geometrical analysis of global stability in trained feedback networks
Francesca Mastrogiuseppe, Srdjan Ostojic
Recurrent neural networks have been extensively studied in the context of neuroscience and machine learning due to their ability to implement complex computations. While substantia…