VC Dimension and Distribution-Free Sample-Based Testing

TL;DR

This paper explores the relationship between VC dimension, a measure of complexity, and the sample complexity of distribution-free property testing, introducing the LVC dimension to derive bounds for various function classes.

Contribution

It introduces the LVC dimension, a variant of VC dimension, to establish tight lower bounds on sample complexity for testing certain classes, and shows some classes can be tested more efficiently than learned.

Findings

LVC dimension helps bound testing sample complexity.

Nearly optimal bounds for unions of intervals, halfspaces, and decision trees.

Juntas and monotone functions can be tested with fewer samples than PAC learning.

Abstract

We consider the problem of determining which classes of functions can be tested more efficiently than they can be learned, in the distribution-free sample-based model that corresponds to the standard PAC learning setting. Our main result shows that while VC dimension by itself does not always provide tight bounds on the number of samples required to test a class of functions in this model, it can be combined with a closely-related variant that we call "lower VC" (or LVC) dimension to obtain strong lower bounds on this sample complexity. We use this result to obtain strong and in many cases nearly optimal lower bounds on the sample complexity for testing unions of intervals, halfspaces, intersections of halfspaces, polynomial threshold functions, and decision trees. Conversely, we show that two natural classes of functions, juntas and monotone functions, can be tested with a number of samples that is polynomially smaller than the number of samples required for PAC learning. Finally, we also use the connection between VC dimension and property testing to establish new lower bounds for testing radius clusterability and testing feasibility of linear constraint systems.