Sharp finite-sample large deviation bounds for independent variables

08/30/2020

∙

by   Akshay Balsubramani, et al.

∙

4

∙

share

We show an extension of Sanov's theorem in large deviations theory, controlling the tail probabilities of i.i.d. random variables with matching concentration and anti-concentration bounds. This result applies to samples of any size, and has a short information-theoretic proof using elementary techniques.