Survey of Multi-Agent Deep RL with GNN Communication
A new survey paper on arXiv reviews multi-agent deep reinforcement learning methods that use graph neural networks for communication. The authors propose a generalized GNN-based communication process to clarify underlying concepts. The paper addresses the lack of explicit structure in classifying such approaches.
Key facts
- The paper is a survey of multi-agent deep reinforcement learning with GNN-based communication.
- It proposes a generalized GNN-based communication process.
- The survey aims to make underlying concepts more accessible.
- The paper is categorized under Computer Science > Machine Learning.
- It was submitted to arXiv.
- The current browse context is machine learning.
- The paper references Semantic Scholar and BibTeX tools.
- arXivLabs is mentioned as a framework for collaborative projects.
Entities
Institutions
- arXiv
- Semantic Scholar
- arXivLabs