Karagözlü, Efe MertYıldırım, Yaşar CahitArarat, ÇagınTekin, CemDasgupta, SMandt, SLi, Y2025-02-282025-02-282024-11-262640-3498https://hdl.handle.net/11693/117032Conference Name: 27th International Conference on Artificial Intelligence and Statistics (AISTATS)Date of Conference: May 02-04, 2024We study pure exploration in bandit problems with vector-valued rewards, where the goal is to (approximately) identify the Pareto set of arms given incomplete preferences induced by a polyhedral convex cone. We address the open problem of designing sampleefficient learning algorithms for such problems. We propose Pareto Vector Bandits (PaVeBa), an adaptive elimination algorithm that nearly matches the gap-dependent and worst-case lower bounds on the sample complexity of (., d)-PAC Pareto set identification. Finally, we provide an in-depth numerical investigation of PaVeBa and its heuristic vari-ants by comparing them with the state-of-the-art multi-objective and vector optimization algorithms on several real-world datasets with conflicting objectives.EnglishOptimizationDesignConference Paper