Cooperative Multi-Agent Systems from the Reinforcement Learning Perspective – Challenges, Algorithms, and an Application

2022/01/26 by Gabel, Thomas
#Multi-agent reinforcement learning #decentralized control #job-shop scheduling

paper · doi:10.4230/dagsemproc.09371.2

Abstract

Reinforcement Learning has established as a framework that allows an autonomous agent for automatically acquiring – in a trial and error-based manner – a behavior policy based on a specification of the desired behavior of the system. In a multi-agent system, however, the decentralization of the control and observation of the system among independent agents has a significant impact on learning and it complexity. In this survey talk, we briefly review the foundations of single-agent reinforcement learning, point to the merits and challenges when applied in a multi-agent setting, and illustrate its potential in the context of an application from the field of manufacturing control and scheduling.

Related

• Deep Reinforcement Learning for Multi-Agent Interaction
• Multi-Agent Reinforcement Learning: A Report on Challenges and Approaches
• A Survey of Multi Agent Reinforcement Learning: Federated Learning and Cooperative and Noncooperative Decentralized Regimes
• Deep Decentralized Reinforcement Learning for Cooperative Control
• Strategy-Following Multi-Agent Deep Reinforcement Learning Considering Control Strategies Provided to Other Agents