A survey of multi-objective sequential decision-making
Version 2 2024-06-04, 15:49Version 2 2024-06-04, 15:49
Version 1 2018-08-24, 14:39Version 1 2018-08-24, 14:39
journal contribution
posted on 2024-06-04, 15:49 authored by DM Roijers, P Vamplew, S Whiteson, Richard DazeleyRichard DazeleySequential decision-making problems with multiple objectives arise naturally in practice and pose unique challenges for research in decision-theoretic planning and learning, which has largely focused on single-objective settings. This article surveys algorithms designed for sequential decision-making problems with multiple objectives. Though there is a growing body of literature on this subject, little of it makes explicit under what circumstances special methods are needed to solve multi-objective problems. Therefore, we identify three distinct scenarios in which converting such a problem to a single-objective one is impossible, infeasible, or undesirable. Furthermore, we propose a taxonomy that classifies multi-objective methods according to the applicable scenario, the nature of the scalarization function (which projects multi-objective values to scalar ones), and the type of policies considered. We show how these factors determine the nature of an optimal solution, which can be a single policy, a convex hull, or a Pareto front. Using this taxonomy, we survey the literature on multi-objective methods for planning and learning. Finally, we discuss key applications of such methods and outline opportunities for future work. © 2013 AI Access Foundation.
History
Journal
Journal of artificial intelligence researchVolume
48Pagination
67-113Location
Palo Alto, Calif.eISSN
1076-9757Language
engPublication classification
C Journal article, C1.1 Refereed article in a scholarly journalCopyright notice
2013, AI Access FoundationPublisher
A A A I PressUsage metrics
Categories
Keywords
Licence
Exports
RefWorksRefWorks
BibTeXBibTeX
Ref. managerRef. manager
EndnoteEndnote
DataCiteDataCite
NLMNLM
DCDC