Informational Confidence Bounds for Self-Normalized Averages and Applications

Aurélien Garivier (1)

1 IMT - Institut de Mathématiques de Toulouse UMR5219

We present deviation bounds for self-normalized averages and applications to estimation with a random number of observations. The results rely on a peeling argument in exponential martingale techniques that represents an alternative to the method of mixture. The motivating examples of bandit problems and context tree estimation are detailed.

Domaines: Statistiques [math.ST]; Théorie [stat.TH]

Aurélien Garivier. Informational Confidence Bounds for Self-Normalized Averages and Applications. IEEE Information Theory Workshop, Sep 2013, Seville, Spain. pp.489-493. ⟨hal-00862062⟩