English

Multi-Objective Reinforcement Learning for Water Management

Machine Learning 2025-11-24 v2 Artificial Intelligence

Abstract

Many real-world problems (e.g., resource management, autonomous driving, drug discovery) require optimizing multiple, conflicting objectives. Multi-objective reinforcement learning (MORL) extends classic reinforcement learning to handle multiple objectives simultaneously, yielding a set of policies that capture various trade-offs. However, the MORL field lacks complex, realistic environments and benchmarks. We introduce a water resource (Nile river basin) management case study and model it as a MORL environment. We then benchmark existing MORL algorithms on this task. Our results show that specialized water management methods outperform state-of-the-art MORL approaches, underscoring the scalability challenges MORL algorithms face in real-world scenarios.

Keywords

Cite

@article{arxiv.2505.01094,
  title  = {Multi-Objective Reinforcement Learning for Water Management},
  author = {Zuzanna Osika and Roxana Rădulescu and Jazmin Zatarain Salazar and Frans Oliehoek and Pradeep K. Murukannaiah},
  journal= {arXiv preprint arXiv:2505.01094},
  year   = {2025}
}

Comments

Accepted to AAMAS 2025

R2 v1 2026-06-28T23:18:57.719Z