Horizontally Scalable Submodular Maximization

[edit]

Mario Lucic, Olivier Bachem, Morteza Zadimoghaddam, Andreas Krause ;
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:2981-2989, 2016.

Abstract

A variety of large-scale machine learning problems can be cast as instances of constrained submodular maximization. Existing approaches for distributed submodular maximization have a critical drawback: The capacity - number of instances that can fit in memory - must grow with the data set size. In practice, while one can provision many machines, the capacity of each machine is limited by physical constraints. We propose a truly scalable approach for distributed submodular maximization under fixed capacity. The proposed framework applies to a broad class of algorithms and constraints and provides theoretical guarantees on the approximation factor for any available capacity. We empirically evaluate the proposed algorithm on a variety of data sets and demonstrate that it achieves performance competitive with the centralized greedy solution.

Related Material