JMLR

Deletion Robust Non-Monotone Submodular Maximization over Matroids

Authors
Paul Dütting Federico Fusco Silvio Lattanzi Ashkan Norouzi-Fard Morteza Zadimoghaddam
Paper Information
  • Journal:
    Journal of Machine Learning Research
  • Added to Tracker:
    Jul 15, 2025
Abstract

We study the deletion robust version of submodular maximization under matroid constraints. The goal is to extract a small-size summary of the data set that contains a high-value independent set even after an adversary deletes some elements. We present constant-factor approximation algorithms, whose space complexity depends on the rank $k$ of the matroid, the number $d$ of deleted elements, and the input precision $\varepsilon$. In the centralized setting we present a $(4.494+O(\varepsilon))$-approximation algorithm with summary size $O( \frac{k+d}{\varepsilon^2}\log \frac{k}{\varepsilon})$ that improves to a $(3.582+O(\varepsilon))$-approximation with $O(k + \frac{d}{\varepsilon^2}\log \frac{k}{\varepsilon})$ summary size when the objective is monotone. In the streaming setting we provide a $(9.294 + O(\varepsilon))$-approximation algorithm with summary size and memory $O(k + \frac{d}{\varepsilon^2}\log \frac{k}{\varepsilon})$; the approximation factor is then improved to $(5.582+O(\varepsilon))$ in the monotone case.

Author Details
Paul Dütting
Author
Federico Fusco
Author
Silvio Lattanzi
Author
Ashkan Norouzi-Fard
Author
Morteza Zadimoghaddam
Author
Citation Information
APA Format
Paul Dütting , Federico Fusco , Silvio Lattanzi , Ashkan Norouzi-Fard & Morteza Zadimoghaddam . Deletion Robust Non-Monotone Submodular Maximization over Matroids. Journal of Machine Learning Research .
BibTeX Format
@article{JMLR:v26:23-1219,
  author  = {Paul D{{\"u}}tting and Federico Fusco and Silvio Lattanzi and Ashkan Norouzi-Fard and Morteza Zadimoghaddam},
  title   = {Deletion Robust Non-Monotone Submodular Maximization over Matroids},
  journal = {Journal of Machine Learning Research},
  year    = {2025},
  volume  = {26},
  number  = {66},
  pages   = {1--28},
  url     = {http://jmlr.org/papers/v26/23-1219.html}
}
Related Papers