Submodular Maximization Subject to Matroid Intersection on the Fly


Date

2022

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric

Data

Abstract

Despite a surge of interest in submodular maximization in the data stream model, there remain significant gaps in our knowledge about what can be achieved in this setting, especially when dealing with multiple constraints. In this work, we nearly close several basic gaps in submodular maximization subject to k matroid constraints in the data stream model. We present a new hardness result showing that super polynomial memory in k is needed to obtain an o(k/log k)-approximation. This implies near optimality of prior algorithms. For the same setting, we show that one can nevertheless obtain a constant-factor approximation by maintaining a set of elements whose size is independent of the stream size. Finally, for bipartite matching constraints, a well-known special case of matroid intersection, we present a new technique to obtain hardness bounds that are significantly stronger than those obtained with prior approaches. Prior results left it open whether a 2-approximation may exist in this setting, and only a complexity-theoretic hardness of 1.91 was known. We prove an unconditional hardness of 2.69.

Publication status

published

Book title

30th Annual European Symposium on Algorithms (ESA 2022)

Volume

244

Pages / Article No.

52

Publisher

Schloss Dagstuhl – Leibniz-Zentrum für Informatik

Event

30th Annual European Symposium on Algorithms (ESA 2022)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Submodular Maximization; matroid intersection; Streaming Algorithms

Organisational unit

09487 - Zenklusen, Rico / Zenklusen, Rico check_circle

Notes

Funding

184622 - Toward Stronger Approximation Algorithms for Fundamental Network Design and Optimization Problems (SNF)
817750 - Fundamental Problems at the Interface of Combinatorial Optimization with Integer Programming and Online Optimization (EC)

Related publications and datasets