Online Hitting Sets for Disks of Bounded Radii
Minati De, Satyam Singh, Csaba D. Tóth
TL;DR
This work advances online hitting-set algorithms for planar geometric objects, achieving $O( ext{log } M ext{log } n)$ competitiveness for disks and positive homothets of convex bodies with radii or scaling in $[1,M]$, by reducing to a finite, lowest-point-property setting and employing layered tilings and line-separated reductions. It also delivers an $O( ext{log } N)$-competitive algorithm for bottomless rectangles in a lattice point setting, via canonical partitions and a structured hitting strategy. The core technique unifies a two-stage approach: reduce to line-separated problems with the lowest-point property, then extend via tilings and multi-layer decompositions to handle radii-varied objects and general convex bodies, including centrally symmetric ones. Collectively, the results push toward $O( ext{log } n)$-competitive online geometric hitting sets in 2D for a broad class of bounded-size range spaces and lay groundwork for higher-dimensional generalizations.
Abstract
We present algorithms for the online minimum hitting set problem in geometric range spaces: given a set $P$ of $n$ points in the plane and a sequence of geometric objects that arrive one-by-one, we need to maintain a hitting set at all times by making irrevocable decisions. For disks of radii in the interval $[1,M]$, we present an $O(\log M \log n)$-competitive algorithm. This result generalizes from disks to positive homothets of any convex body in the plane with scaling factors in the interval $[1,M]$. As a main technical tool, we reduce the problem to the online hitting set problem for a finite subset of integer points and geometric objects with the lowest point property, introduced in this paper, which behave similarly to bottomless rectangles. Specifically, for a given $N>1$, we present an $O(\log N)$-competitive algorithm for the variant where $P$ is a subset of an $N\times N$ section of the integer lattice, and the geometric objects have the lowest point property.
