Streaming k-Submodular Maximization under Noise subject to Size Constraint

Maximizing on k-submodular functions subject to size constraint has received extensive attention recently. In this paper, we investigate a more realistic scenario of this problem that (1) obtaining exact evaluation of an objective function is impractical, instead, its noisy version is acquired; and (2) algorithms are required to take only one single pass over dataset, producing solutions in a timely manner. We propose two novel streaming algorithms, namely DSTREAM and RSTREAM, with their theoretical performance guarantees. We further demonstrate the efficiency of our algorithms in two applications in Influence Maximization and Sensor Placement, showing that our algorithms can return comparative results to state-of-the-art non-streaming methods while using a much fewer number of queries.