Computer Science > Data Structures and Algorithms
[Submitted on 29 Sep 2020 (this version), latest version 13 Jul 2021 (v2)]
Title:Simultaneous Greedys: A Swiss Army Knife for Constrained Submodular Maximization
View PDFAbstract:In this paper, we present SimultaneousGreedys, a deterministic algorithm for constrained submodular maximization. At a high level, the algorithm maintains $\ell$ solutions and greedily updates them in a simultaneous fashion, rather than a sequential one. SimultaneousGreedys achieves the tightest known approximation guarantees for both $k$-extendible systems and the more general $k$-systems, which are $(k+1)^2/k = k + \mathcal{O}(1)$ and $(1 + \sqrt{k+2})^2 = k + \mathcal{O}(\sqrt{k})$, respectively. This is in contrast to previous algorithms, which are designed to provide tight approximation guarantees in one setting, but not both. Furthermore, these approximation guarantees further improve to $k+1$ when the objective is monotone. We demonstrate that the algorithm may be modified to run in nearly linear time with an arbitrarily small loss in the approximation. This leads to the first nearly linear time algorithm for submodular maximization over $k$-extendible systems and $k$-systems. Finally, the technique is flexible enough to incorporate the intersection of $m$ additional knapsack constraints, while retaining similar approximation guarantees, which are roughly $k + 2m + \mathcal{O}(\sqrt{k+m})$ for $k$-systems and $k+2m + \mathcal{O}(\sqrt{m})$ for $k$-extendible systems. To complement our algorithmic contributions, we provide a hardness result which states that no algorithm making polynomially many queries to the value and independence oracles can achieve an approximation better than $k + 1/2 + \varepsilon$.
Submission history
From: Christopher Harshaw [view email][v1] Tue, 29 Sep 2020 13:34:09 UTC (57 KB)
[v2] Tue, 13 Jul 2021 18:02:50 UTC (998 KB)
Current browse context:
cs.DS
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.