Locally generalised multi-agent reinforcement learning for demand and capacity balancing with customised neural networks

General Multi-Agent Reinforcement Learning Integrating Heuristic-Based Delay Priority Strategy for Demand and Capacity Balancing

“…The three state-of-the-art methods were selected because they are all capable of solving unseen problems in specific situations. They are named based on their technical highlights: MARL_HC (Huang and Xu, 2021), MARL_CN (Chen et al, 2023) and MARL_AS (Tang and Xu, 2021), where HC, CN and AS refer to Heuristic Cluster, Customised Network and Action Supervisor, respectively.…”