output
20022026
most citedFast learning rates for plug-in classifiers

332 citations

Showing 2021Show all

17 papers · 1 filter

math.ST2021★ 2 cited

Generalized Pareto Regression Trees for extreme events analysis

Sébastien Farkas, Antoine Heranval, Olivier Lopez +1

In this paper, we provide finite sample results to assess the consistency of Generalized Pareto regression trees, as tools to perform extreme value regression. The results that we…

math.PR2021★ 8 cited

Minimal bipartite dimers and higher genus Harnack curves

Cédric Boutillier, David Cimasoni, Béatrice de Tilière

This paper completes the comprehensive study of the dimer model on infinite minimal graphs with Fock's weights [arXiv:1503.00289] initiated in [arXiv:2007.14699]: the latter articl…

stat.ML2021★ 2 cited

Model-based Clustering with Missing Not At Random Data

Aude Sportisse, Matthieu Marbac, Fabien Laporte +4

Model-based unsupervised learning, as any learning task, stalls as soon as missing data occurs. This is even more true when the missing data are informative, or said missing not at…

q-fin.PR2021★ 6 cited

CBI-time-changed Lévy processes for multi-currency modeling

Claudio Fontana, Alessandro Gnoatto, Guillaume Szulda

We develop a stochastic volatility framework for modeling multiple currencies based on CBI-time-changed Lévy processes. The proposed framework captures the typical risk characteris…

cond-mat.stat-mech2021

Macroscopic stability of time evolution of Gibbs measures

Raphael Lefevere, Shin-ichi Sasa

We introduce two properties, macroscopic mixing and transitive mixing, to represent the macroscopic stability of time evolution of Gibbs measures. We claim that these are fundament…

math.NA2021

Strong L2 convergence of time Euler schemes for stochastic 3D Brinkman-Forchheimer-Navier-Stokes equations

Hakima Bessaih, Annie Millet

We prove that some time Euler schemes for the 3D Navier-Stokes equations modified by adding a Brinkman-Forchheimer term and a random perturbation converge in . This extends…