chaosddp / maro Goto Github PK
View Code? Open in Web Editor NEWThis project forked from microsoft/maro
MARO (Multi-Agent Resource Optimization) serves as a domain-specific Reinforcement Learning (RL) solution. In MARO, multi-agent RL is used to solve real-world resource optimization problems. It can be applied to many important industrial domains, such as empty container repositioning (ECR) in logistics, bike repositioning in transportation, virtual machine (VM) provisioning in data centers, and asset management in finance, etc. Besides, MARO is not limited to RL, we also support other planning/decision-related components, such as Operations Research (OR) based planning, which is already in our further worklist. MARO provides comprehensive support in data processing, simulator building, RL algorithm selection, and distributed training.
License: MIT License