Novel Deviation Bounds for Mixture of Independent Bernoulli Variables with Application to the Missing Mass
This work addresses a foundational issue in learning theory by providing improved bounds for missing mass, which is crucial for connecting sample-based density estimates to populations in discrete distributions, though it is incremental in refining existing inequalities.
The paper tackles the problem of deriving concentration inequalities for mixtures of independent Bernoulli variables, specifically applying them to the missing mass in discrete distributions, resulting in Bernstein-like bounds with exponents that behave almost linearly with respect to deviation size and sharpening prior results for large sample sizes.
In this paper, we are concerned with obtaining distribution-free concentration inequalities for mixture of independent Bernoulli variables that incorporate a notion of variance. Missing mass is the total probability mass associated to the outcomes that have not been seen in a given sample which is an important quantity that connects density estimates obtained from a sample to the population for discrete distributions. Therefore, we are specifically motivated to apply our method to study the concentration of missing mass - which can be expressed as a mixture of Bernoulli - in a novel way. We not only derive - for the first time - Bernstein-like large deviation bounds for the missing mass whose exponents behave almost linearly with respect to deviation size, but also sharpen McAllester and Ortiz (2003) and Berend and Kontorovich (2013) for large sample sizes in the case of small deviations which is the most interesting case in learning theory. In the meantime, our approach shows that the heterogeneity issue introduced in McAllester and Ortiz (2003) is resolvable in the case of missing mass in the sense that one can use standard inequalities but it may not lead to strong results. Thus, we postulate that our results are general and can be applied to provide potentially sharp Bernstein-like bounds under some constraints.