Streaming algorithms for maximizing the difference of submodular functions and the sum of submodular and supermodular functions

被引:0
作者
Cheng Lu
Wenguo Yang
Suixiang Gao
机构
[1] University of Chinese Academy of Sciences,School of Mathematical Sciences
来源
Optimization Letters | 2023年 / 17卷
关键词
Nonsubmodular maximization; Difference of submodular functions; Sum of submodular and supermodular functions; Streaming algorithm; Curvature;
D O I
暂无
中图分类号
学科分类号
摘要
In this paper, we study the problem of maximizing the Difference of two Submodular (DS) functions in the streaming model, where elements in the ground set arrive one at a time in an arbitrary order. We present one-pass streaming algorithms for both the unconstrained and cardinality-constrained problems. Our analysis shows that the algorithms we propose are able to produce solutions with provable approximation guarantees. To the best of our knowledge, this is the first theoretical guarantee for the DS maximization problem in the streaming model. In addition, we study the function maximization problem under a cardinality constraint, where the underlying objective function is a γ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma $$\end{document}-weakly DR-submodular function, in the streaming setting. We propose a one-pass streaming algorithm, which achieves an approximation ratio of γ/(1+γ)-ϵ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma /(1 + \gamma ) - \epsilon $$\end{document}. Since the sum of suBmodular and suPermodular (BP) functions can be regarded as a (1-κg)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(1 - \kappa ^g)$$\end{document}-weakly DR-submodular function, we obtain a ((1-κg)/(2-κg)-ϵ)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$( (1 - \kappa ^g)/(2 - \kappa ^g) - \epsilon )$$\end{document}-approximation for the cardinality-constrained BP maximization, where κg\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\kappa ^g$$\end{document} is the curvature of the corresponding supermodular function. Our results improve the previous best approximation bounds.
引用
收藏
页码:1643 / 1667
页数:24
相关论文
共 59 条
  • [1] Buchbinder N(2019)Online submodular maximization with preemption ACM Trans. Algorithms (TALG) 15 1-31
  • [2] Feldman Moran(2015)A tight analysis of the submodular-supermodular procedure Discret. Appl. Math. 186 275-282
  • [3] Schwartz R(2011)Maximizing a monotone submodular function subject to a matroid constraint SIAM J. Comput. 40 1740-1766
  • [4] Byrnes Kevin M(1984)Submodular set functions, matroids and the greedy algorithm: tight worst-case bounds and some generalizations of the rado-edmonds theorem Discret. Appl. Math. 7 251-274
  • [5] Calinescu G(2018)Approximate submodularity and its applications: Subset selection, sparse approximation and dictionary selection J. Mach. Learn. Res. 19 74-107
  • [6] Chekuri Chandra(2013)Pass approximation: a framework for analyzing and designing heuristics Algorithmica 66 450-478
  • [7] Pal M(2021)Guess free maximization of submodular and linear sums Algorithmica 83 853-878
  • [8] Vondrák J(2021)Maximize a monotone function with a generic submodularity ratio Theor. Comput. Sci. 853 16-24
  • [9] Conforti M(2021)Unconstrained submodular maximization with modular costs: tight approximation and application to profit maximization Proc. VLDB Endow. 14 1756-1768
  • [10] Cornuéjols Gérard(2008)Near-optimal sensor placements in gaussian processes: theory, efficient algorithms and empirical studies J. Mach. Learn. Res. 9 235-284