Multi-Agent Systems @ Birmingham (MAS@B)
Welcome to MAS@B! We are an interdisciplinary interest group spanning research themes across the School of Computer Science at the University of Birmingham.
Multi-Agent Systems (MAS) is fundamentally cross-cutting. We bring together researchers, faculty, and PhD students working on theoretical foundations, algorithmic frameworks, hardware deployments, and emerging AI models where autonomous entities interact.
Broad Research Themes
- MARL: Multi-Agent Reinforcement Learning, Human-Agent Teaming, communication protocols.
- Robotics & Swarms: Multi-robot coordination, swarm systems, multi-agent pathfinding (MAPF).
- MAS for IoT, Tiny Devices & Edge AI: Resource-constrained multi-agent execution, TinyML, dynamic sensor networks, decentralized edge computing.
- Game Theory & Mechanism Design: Strategic decision-making, dynamic games, auction design, fair resource allocation.
- Agentic AI & LLM Frameworks: Multi-agent LLM systems, autonomous workflows, tool-use orchestration.
What We Do
- Informal Knowledge Exchange: Discussions to share work-in-progress, test early-stage ideas, and troubleshoot research roadblocks in a relaxed environment.
- Student & Researcher Tutorials: Hands-on deep dives led by PhD students and postdocs to cross-develop technical skills across labs.
- Cross-Group Funding & Proposals: Identifying grant opportunities and assembling joint research bids.
- Workshops & Event Signposting: Organizing internal one-day workshops and cross-promoting MAS-related seminars across campus.