Title: An Optimal Streaming Algorithm for Non-monotone Submodular Maximization

Abstract: We study the problem of maximizing a non-monotone submodular function subject to a size constraint in the streaming model. Our main contribution is a single-pass streaming algorithm that uses $k\cdot\mathrm{poly}(1/\epsilon)$ memory, where $k$ is the size constraint. At the end of the stream, we post-process the output of the algorithm using any offline algorithm for submodular maximization, and we obtain a solution whose approximation guarantee is $\frac{\alpha}{1+\alpha}-\epsilon$, where $\alpha$ is the approximation of the offline algorithm. If we use an exact (exponential time) post-processing algorithm, we obtain a $\frac{1}{2}-\epsilon$ approximation, almost matching the lower bound of [Alaluf-Feldman, 2019]. If we post-process with the algorithm of [Buchbinder-Feldman, Math of OR 2019] that achieves the currently best approximation guarantee $\alpha=0.385$, we obtain a $0.2779$ approximation in polynomial time, improving over the previously best polynomial-time approximation of $0.2335$ due to [Alaluf-Feldman, 2019]. In addition to its improved approximation guarantee, our algorithm enjoys a fast update time and overall running time.