Rencontres Statistiques du CEREMADE (Quentin Duchemin, lundi 24 mars 2025)

La prochaine séance des Rencontres statistiques du CEREMADE aura lieu le lundi 24 mars 2025 à 13h30 en salle P428. Nous aurons le plaisir d'écouter Quentin Duchemin (EPFL), qui nous parlera de

Pinball Quantile Regression Trees for Conformal Prediction

Abstract
Conformal prediction, known for providing model-agnostic uncertainty quantification, has become an essential tool with a mushrooming number of applications. An ideal conformal prediction method should be adaptive, data-efficient, and interpretable. To meet these criteria, we propose to improve existing quantile-based conformal methods using Pinball quantile regression trees. Moreover, we introduce a new quantile out-of-bag framework that is shown empirically to provide smaller conformal prediction sets while maintaining rigorous theoretical guarantees for coverage.

Page web : http://quentin-duchemin.alwaysdata.net/wiki/