Title: A QPTAS for stabbing rectangles

Abstract: We consider the following geometric optimization problem: Given $ n $ axis-aligned rectangles in the plane, the goal is to find a set of horizontal segments of minimum total length such that each rectangle is stabbed. A segment stabs a rectangle if it intersects both its left and right edge. As such, this stabbing problem falls into the category of weighted geometric set cover problems for which techniques that improve upon the general ${\Theta}(\log n)$-approximation guarantee have received a lot of attention in the literature. Chan at al. (2018) have shown that rectangle stabbing is NP-hard and that it admits a constant-factor approximation algorithm based on Varadarajan's quasi-uniform sampling method. In this work we make progress on rectangle stabbing on two fronts. First, we present a quasi-polynomial time approximation scheme (QPTAS) for rectangle stabbing. Furthermore, we provide a simple $8$-approximation algorithm that avoids the framework of Varadarajan. This settles two open problems raised by Chan et al. (2018).