Robust Adaptive Submodular Maximization

The goal of a sequential decision-making problem is to design an interactive policy that adaptively selects group items, each selection based on the feedback from past, maximize expected utility selected items. It has been shown functions many real-world applications are adaptive submodular. However, most existing studies submodular optimization focus average-case, is, their objective find maximizes over known distribution realizations. Unfortunately, good average-case performance may have very poor under worst-case realization. In this study, we propose study two variants problems, namely, maximization and robust maximization. first aims latter one policy, if any, achieves both near optimal simultaneously. We introduce new class stochastic functions, called function. For subject p-system constraint, develop greedy [Formula: see text] approximation ratio against function cardinality constraints (respectively, partition matroid constraints), submodular, hybrid close 1/3) worst- settings also describe several our theoretical results, including pool-base active learning, set cover, viral marketing. History: Accepted by Andrea Lodi, Area Editor for Design & Analysis Algorithms—Discrete. Supplemental Material: online appendix available at https://doi.org/10.1287/ijoc.2022.1239 .