Scalable Deletion-Robust Submodular Maximization: Data Summarization with Privacy and Fairness Constraints

TL;DR: This work proposes the first memory-efficient centralized, streaming, and distributed methods with constant-factor approximation guarantees against any number of adversarial deletions, and shows that the solution is robust against even 80% of data deletion.

Abstract: Can we efficiently extract useful information from a large user-generated dataset while protecting the privacy of the users and/or ensuring fairness in representation? We cast this problem as an instance of a deletion-robust submodular maximization where part of the data may be deleted or masked due to privacy concerns or fairness criteria. We propose the first memory-efficient centralized, streaming, and distributed methods with constant-factor approximation guarantees against any number of adversarial deletions. We extensively evaluate the performance of our algorithms on real-world applications, including (i) Uber-pick up locations with location privacy constraints; (ii) feature selection with fairness constraints for income prediction and crime rate prediction; and (iii) robust to deletion summarization of census data, consisting of 2,458,285 feature vectors. Our experiments show that our solution is robust against even 80% of data deletion.