We study the Steiner Tree problem on the intersection graph of most natural families of geometric objects, e.g., disks, squares, polygons, etc. Given a set of $n$ objects in the plane and a subset $T$ of $t$ terminal objects, the task is to find a subset $S$ of $k$ objects such that the intersection graph of $S\cup T$ is connected. Given how typical parameterized problems behave on planar graphs and geometric intersection graphs, we would expect that exact algorithms with some form of subexponential dependence on the solution size or the number of terminals exist. Contrary to this expectation, we show that, assuming the Exponential-Time Hypothesis (ETH), there is no $2^{o(k+t)}\cdot n^{O(1)}$ time algorithm even for unit disks or unit squares, that is, there is no FPT algorithm subexponential in the size of the Steiner tree. However, subexponential dependence can appear in a different form: we show that Steiner Tree can be solved in time $n^{O(\sqrt{t})}$ for many natural classes of objects, including: Disks of arbitrary size. Axis-parallel squares of arbitrary size. Similarly-sized fat polygons. This in particular significantly improves and generalizes two recent results: (1) Steiner Tree on unit disks can be solved in time $n^{\Oh(\sqrt{k + t})}$ (Bhore, Carmi, Kolay, and Zehavi, Algorithmica 2023) and (2) Steiner Tree on planar graphs can be solved in time $n^{O(\sqrt{t})}$ (Marx, Pilipczuk, and Pilipczuk, FOCS 2018). We complement our algorithms with lower bounds that demonstrate that the class of objects cannot be significantly extended, even if we allow the running time to be $n^{o(k+t)/\log(k+t)}$.
翻译:我们研究了在大多数自然几何对象族(如圆盘、正方形、多边形等)的交图上求解Steiner树问题。给定平面上n个对象的集合以及t个终端对象的子集T,任务是找到k个对象的子集S,使得S∪T的交图连通。鉴于参数化问题在平面图和几何交图上的典型表现,我们预期存在某种形式的、关于解大小或终端数量的亚指数依赖的精确算法。然而,与这一预期相反,我们证明:假设指数时间假说(ETH)成立,即使对于单位圆盘或单位正方形,也不存在运行时间为2^{o(k+t)}·n^{O(1)}的算法,即不存在关于Steiner树大小的亚指数FPT算法。但是,亚指数依赖可以以不同形式出现:我们证明对于许多自然对象类,Steiner树问题可以在n^{O(√t)}时间内求解,包括:任意大小的圆盘、任意大小的轴平行正方形、相似大小的胖多边形。这特别显著改进并推广了最近的两项结果:(1)单位圆盘上的Steiner树问题可在n^{O(√k+t)}时间内求解(Bhore、Carmi、Kolay和Zehavi,Algorithmica 2023);(2)平面图上的Steiner树问题可在n^{O(√t)}时间内求解(Marx、Pilipczuk和Pilipczuk,FOCS 2018)。我们通过下界分析对算法进行补充,证明即使允许运行时间为n^{o(k+t)/log(k+t)},对象类的范围也无法显著扩展。