An Asymptotic Law of the Iterated Logarithm for $\mathrm{KL}_{\inf}$

TL;DR

This paper establishes a precise asymptotic law of the iterated logarithm for the empirical _{ ext{inf}} statistic, applicable to unbounded data, which is crucial in bandit algorithms and sequential testing.

Contribution

It provides the first tight law of the iterated logarithm for empirical _{ ext{inf}} applicable to general unbounded data, improving understanding of its asymptotic fluctuations.

Findings

Derives a tight law of the iterated logarithm for empirical _{ ext{inf}}.

Applicable to a broad class of unbounded data.

Enhances theoretical understanding of _{ ext{inf}} fluctuations.

Abstract

The population $\mathrm{KL}_{\inf}$ is a fundamental quantity that appears in lower bounds for (asymptotically) optimal regret of pure-exploration stochastic bandit algorithms, and optimal stopping time of sequential tests. Motivated by this, an empirical $\mathrm{KL}_{\inf}$ statistic is frequently used in the design of (asymptotically) optimal bandit algorithms and sequential tests. While nonasymptotic concentration bounds for the empirical $\mathrm{KL}_{\inf}$ have been developed, their optimality in terms of constants and rates is questionable, and their generality is limited (usually to bounded observations). The fundamental limits of nonasymptotic concentration are often described by the asymptotic fluctuations of the statistics. With that motivation, this paper presents a tight (upper and lower) law of the iterated logarithm for empirical $\mathrm{KL}_{\inf}$ applying to extremely general (unbounded) data.