Testing Identity of Multidimensional Histograms

We investigate the problem of identity testing for multidimensional histogram distributions. A distribution , where , is called a -histogram if there exists a partition of the domain into axis-aligned rectangles such that is constant within each such rectangle. Histograms are one of the most fundamental nonparametric families of distributions and have been extensively studied in computer science and statistics. We give the first identity tester for this problem with {\em sub-learning} sample complexity in any fixed dimension and a nearly-matching sample complexity lower bound. In more detail, let be an unknown -dimensional -histogram distribution in fixed dimension , and be an explicitly given -dimensional -histogram. We want to correctly distinguish, with probability at least , between the case that versus . We design an algorithm for this hypothesis testing problem with sample complexity that runs in sample-polynomial time. Our algorithm is robust to model misspecification, i.e., succeeds even if is only promised to be {\em close} to a -histogram. Moreover, for , we show a sample complexity lower bound of when . That is, for any fixed dimension , our upper and lower bounds are nearly matching. Prior to our work, the sample complexity of the case was well-understood, but no algorithm with sub-learning sample complexity was known, even for . Our new upper and lower bounds have interesting conceptual implications regarding the relation between learning and testing in this setting.
View on arXiv