preprint · arXiv (Cornell University) · 2020

Submodular Maximization Through Barrier Functions

Ashwinkumar Badanidiyuru, Amin Karbasi, Ehsan Kazemi, J. Vondrák

Cited by 3Open access

Abstract

In this paper, we introduce a novel technique for constrained submodular maximization, inspired by barrier functions in continuous optimization. This connection not only improves the running time for constrained submodular maximization but also provides the state of the art guarantee. More precisely, for maximizing a monotone submodular function subject to the combination of a $k$-matchoid and $\ell$-knapsack constraint (for $\ell\leq k$), we propose a potential function that can be approximately minimized. Once we minimize the potential function up to an $ε$ error it is guaranteed that we have found a feasible set with a $2(k+1+ε)$-approximation factor which can indeed be further improved to $(k+1+ε)$ by an enumeration technique. We extensively evaluate the performance of our proposed algorithm over several real-world applications, including a movie recommendation system, summarization tasks for YouTube videos, Twitter feeds and Yelp business locations, and a set cover problem.

Cite this paper

APA 7

Badanidiyuru, A., Karbasi, A., Kazemi, E., & Vondrák, J. (2020). Submodular Maximization Through Barrier Functions. arXiv (Cornell University), 33, 524–534. https://doi.org/10.48550/arxiv.2002.03523

MLA 9

Badanidiyuru, Ashwinkumar, et al. “Submodular Maximization Through Barrier Functions.” arXiv (Cornell University), vol. 33, 2020, pp. 524–34, https://doi.org/10.48550/arxiv.2002.03523.

Chicago

Badanidiyuru, Ashwinkumar, Amin Karbasi, Ehsan Kazemi, and J. Vondrák. 2020. “Submodular Maximization Through Barrier Functions.” arXiv (Cornell University) 33: 524–34. https://doi.org/10.48550/arxiv.2002.03523.

Harvard

Badanidiyuru, A. et al. (2020) ‘Submodular Maximization Through Barrier Functions’, arXiv (Cornell University), 33, pp. 524–534. Available at: https://doi.org/10.48550/arxiv.2002.03523.

BibTeX

@misc{badanidiyuru2020submodular,
  title = {{Submodular Maximization Through Barrier Functions}},
  author = {Badanidiyuru, Ashwinkumar and Karbasi, Amin and Kazemi, Ehsan and Vondrák, J.},
  journal = {arXiv (Cornell University)},
  volume = {33},
  pages = {524--534},
  year = {2020},
  doi = {10.48550/arxiv.2002.03523}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.