We consider the problem of solving the Min-Sum Submodular Cover problem using local search. The Min-Sum Submodular Cover problem generalizes the NP-complete Min-Sum Set Cover problem, replacing the input set cover instance with a monotone submodular set function. A simple greedy algorithm achieves an approximation factor of 4, which is tight unless P=NP [Streeter and Golovin, NeurIPS, 2008]. We complement the greedy algorithm with analysis of a local search algorithm. Building on work of Munagala et al. [ICDT, 2005], we show that, using simple initialization, a straightforward local search algorithm achieves a $(4+\epsilon)$-approximate solution in time $O(n^3\log(n/\epsilon))$, provided that the monotone submodular set function is also second-order supermodular. Second-order supermodularity has been shown to hold for a number of submodular functions of practical interest, including functions associated with set cover, matching, and facility location. We present experiments on two special cases of Min-Sum Submodular Cover and find that the local search algorithm can outperform the greedy algorithm on small data sets.
翻译:我们考虑使用本地搜索解决 Min-Sum 子模块覆盖问题。 Min-Sum 子模块覆盖问题将NP-complete Min-Sum-Set 覆盖问题普遍化,用单调子模块功能取代输入集覆盖实例。简单贪婪算法的近似系数为4,除非P=NP[Streeter和Golovin, NeurIPS,2008]。我们用对本地搜索算法的分析来补充贪婪算法。根据Munagala等人的工作[ICDT, 2005],我们显示,使用简单的初始化,直截了当的本地搜索算法在时间上实现了$(4 ⁇ -epsilon)$的近似解决方案。只要单调子组合函数也是二级超级模式,那么单调子模块的设置功能就很紧凑。二级超级模式已经显示,用于一系列具有实际兴趣的子模块功能,包括与设定覆盖、匹配和设施位置相关的功能。我们介绍了两个关于软调的本地数据系统搜索和小型版本的特殊案例。