Demonstration Guided Multi-Objective Reinforcement Learning

arXiv:2404.03997v1 Announce Type: new
Abstract: Multi-objective reinforcement learning (MORL) is increasingly relevant due to its resemblance to real-world scenarios requiring trade-offs between multiple objectives. Catering to diverse user preferences, traditional reinforcement learning faces amplified challenges in MORL. To address the difficulty of training policies from scratch in MORL, we introduce demonstration-guided multi-objective reinforcement learning (DG-MORL). This novel approach utilizes prior demonstrations, aligns them with user preferences via corner weight support, and incorporates a self-evolving mechanism to refine suboptimal demonstrations. Our …

abstract arxiv challenges cs.ai cs.lg diverse multi-objective multiple policies reinforcement reinforcement learning scratch trade training type world