Sample-based high-dimensional convexity testing
Abstract
In the problem of high-dimensional convexity testing, there is an unknown set which is promised to be either convex or -far from every convex body with respect to the standard multivariate normal distribution . The job of a testing algorithm is then to distinguish between these two cases while making as few inspections of the set as possible. In this work we consider sample-based testing algorithms, in which the testing algorithm only has access to labeled samples where each is independently drawn from . We give nearly matching sample complexity upper and lower bounds for both one-sided and two-sided convexity testing algorithms in this framework. For constant , our results show that the sample complexity of one-sided convexity testing is samples, while for two-sided convexity testing it is .
Cite
@article{arxiv.1706.09362,
title = {Sample-based high-dimensional convexity testing},
author = {Xi Chen and Adam Freilich and Rocco A. Servedio and Timothy Sun},
journal= {arXiv preprint arXiv:1706.09362},
year = {2017}
}