AI RESEARCH
Constrained Multi-Objective Reinforcement Learning with Max-Min Criterion
arXiv CS.LG
•
ArXi:2605.31388v1 Announce Type: new Multi-Objective Reinforcement Learning (MORL) extends standard RL by optimizing policies with respect to multiple, often conflicting, objectives. While max-min MORL has emerged as an effective approach for promoting fairness, its applicability remains limited, particularly when constraints must be incorporated. In this paper, we propose a MORL framework that integrates the max-min criterion with explicit constraint satisfaction.