English

Robust Submodular Maximization: A Non-Uniform Partitioning Approach

Machine Learning 2017-06-16 v1 Machine Learning

Abstract

We study the problem of maximizing a monotone submodular function subject to a cardinality constraint kk, with the added twist that a number of items τ\tau from the returned set may be removed. We focus on the worst-case setting considered in (Orlin et al., 2016), in which a constant-factor approximation guarantee was given for τ=o(k)\tau = o(\sqrt{k}). In this paper, we solve a key open problem raised therein, presenting a new Partitioned Robust (PRo) submodular maximization algorithm that achieves the same guarantee for more general τ=o(k)\tau = o(k). Our algorithm constructs partitions consisting of buckets with exponentially increasing sizes, and applies standard submodular optimization subroutines on the buckets in order to construct the robust solution. We numerically demonstrate the performance of PRo in data summarization and influence maximization, demonstrating gains over both the greedy algorithm and the algorithm of (Orlin et al., 2016).

Keywords

Cite

@article{arxiv.1706.04918,
  title  = {Robust Submodular Maximization: A Non-Uniform Partitioning Approach},
  author = {Ilija Bogunovic and Slobodan Mitrović and Jonathan Scarlett and Volkan Cevher},
  journal= {arXiv preprint arXiv:1706.04918},
  year   = {2017}
}

Comments

Accepted to ICML 2017

R2 v1 2026-06-22T20:19:51.728Z