Non-Smooth, H\"older-Smooth, and Robust Submodular Maximization
Abstract
We study the problem of maximizing a continuous DR-submodular function that is not necessarily smooth. We prove that the continuous greedy algorithm achieves an guarantee when the function is monotone and H\"older-smooth, meaning that it admits a H\"older-continuous gradient. For functions that are non-differentiable or non-smooth, we propose a variant of the mirror-prox algorithm that attains an guarantee. We apply our algorithmic frameworks to robust submodular maximization and distributionally robust submodular maximization under Wasserstein ambiguity. In particular, the mirror-prox method applies to robust submodular maximization to obtain a single feasible solution whose value is at least . For distributionally robust maximization under Wasserstein ambiguity, we deduce and work over a submodular-convex maximin reformulation whose objective function is H\"older-smooth, for which we may apply both the continuous greedy and the mirror-prox algorithms.
Cite
@article{arxiv.2210.06061,
title = {Non-Smooth, H\"older-Smooth, and Robust Submodular Maximization},
author = {Duksang Lee and Nam Ho-Nguyen and Dabeen Lee},
journal= {arXiv preprint arXiv:2210.06061},
year = {2023}
}