Mixture Martingales Revisited with Applications to Sequential Tests and Confidence Intervals

This paper presents new deviation inequalities that are valid uniformly in time under adaptive sampling in a multi-armed bandit model. The deviations are measured using the Kullback-Leibler divergence in a given one-dimensional exponential family, and may take into account several arms at a time. Th...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of machine learning research 2021-12
Hauptverfasser:	Kaufmann, Emilie, Koolen, Wouter M.
Format:	Artikel
Sprache:	eng
Schlagworte:	Machine Learning Statistics
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!