Deletion Robust Non-Monotone Submodular Maximization over Matroids

Maximizing a submodular function is a fundamental task in machine learning and in this paper we study the deletion robust version of the problem under the classic matroids constraint. Here the goal is to extract a small size summary of the dataset that contains a high value independent set even after an adversary deleted some elements. We present constant-factor approximation algorithms, whose space complexity depends on the rank of the matroid and the number of deleted elements. In the centralized setting we present a -approximation algorithm with summary size that is improved to a -approximation with summary size when the objective is monotone. In the streaming setting we provide a -approximation algorithm with summary size and memory ; the approximation factor is then improved to in the monotone case.
View on arXiv