Optimistic Dynamic Regret Bounds - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2023

Optimistic Dynamic Regret Bounds

Résumé

Online Learning (OL) algorithms have originally been developed to guarantee good performances when comparing their output to the best fixed strategy. The question of performance with respect to dynamic strategies remains an active research topic. We develop in this work dynamic adaptations of classical OL algorithms based on the use of experts' advice and the notion of optimism. We also propose a constructivist method to generate those advices and eventually provide both theoretical and experimental guarantees for our procedures.

Dates et versions

hal-03953310 , version 1 (23-01-2023)

Identifiants

Citer

Maxime Haddouche, Benjamin Guedj, Olivier Wintenberger. Optimistic Dynamic Regret Bounds. 2023. ⟨hal-03953310⟩
27 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More