English

Robust Monotone Submodular Function Maximization

Data Structures and Algorithms 2018-10-31 v4 Discrete Mathematics Optimization and Control

Abstract

We consider a robust formulation, introduced by Krause et al. (2008), of the classical cardinality constrained monotone submodular function maximization problem, and give the first constant factor approximation results. The robustness considered is w.r.t. adversarial removal of up to τ\tau elements from the chosen set. For the fundamental case of τ=1\tau=1, we give a deterministic (11/e)1/Θ(m)(1-1/e)-1/\Theta(m) approximation algorithm, where mm is an input parameter and number of queries scale as O(nm+1)O(n^{m+1}). In the process, we develop a deterministic (11/e)1/Θ(m)(1-1/e)-1/\Theta(m) approximate greedy algorithm for bi-objective maximization of (two) monotone submodular functions. Generalizing the ideas and using a result from Chekuri et al. (2010), we show a randomized (11/e)ϵ(1-1/e)-\epsilon approximation for constant τ\tau and ϵ1Ω~(τ)\epsilon\leq \frac{1}{\tilde{\Omega}(\tau)}, making O(n1/ϵ3)O(n^{1/\epsilon^3}) queries. Further, for τk\tau\ll \sqrt{k}, we give a fast and practical 0.387 algorithm. Finally, we also give a black box result result for the much more general setting of robust maximization subject to an Independence System.

Keywords

Cite

@article{arxiv.1507.06616,
  title  = {Robust Monotone Submodular Function Maximization},
  author = {James B. Orlin and Andreas S. Schulz and Rajan Udwani},
  journal= {arXiv preprint arXiv:1507.06616},
  year   = {2018}
}

Comments

Preliminary version in IPCO 2016

R2 v1 2026-06-22T10:17:23.717Z