Tighter estimates for ϵ-nets for disks

The geometric hitting set problem is one of the basic geometric combinatorial optimization problems: given a set P   of points, and a set DD of geometric objects in the plane, the goal is to compute a small-sized subset of P   that hits all objects in DD. In 1994, Bronnimann and Goodrich [5] made an important connection of this problem to the size of fundamental combinatorial structures called ϵ-nets, showing that small-sized ϵ-nets imply approximation algorithms with correspondingly small approximation ratios. Very recently, Agarwal and Pan [2] showed that their scheme can be implemented in near-linear time for disks in the plane. Altogether this gives O(1)O(1)-factor approximation algorithms in O˜(n) time for hitting sets for disks in the plane.This constant factor depends on the sizes of ϵ  -nets for disks; unfortunately, the current state-of-the-art bounds are large – at least 24/ϵ24/ϵ and most likely larger than 40/ϵ40/ϵ. Thus the approximation factor of the Agarwal and Pan algorithm ends up being more than 40. The best lower-bound is 2/ϵ2/ϵ, which follows from the Pach–Woeginger construction [32] for halfplanes in two dimensions. Thus there is a large gap between the best-known upper and lower bounds. Besides being of independent interest, finding precise bounds is important since this immediately implies an improved linear-time algorithm for the hitting-set problem.The main goal of this paper is to improve the upper-bound to 13.4/ϵ13.4/ϵ for disks in the plane. The proof is constructive, giving a simple algorithm that uses only Delaunay triangulations. We have implemented the algorithm, which is available as a public open-source module. Experimental results show that the sizes of ϵ  -nets for a variety of data-sets are lower, around 9/ϵ9/ϵ.