AAISI

The Multi-Agent AI Reading Group

This is a collaborative forum dedicated to understanding and addressing the unique challenges posed by multi-agent AI systems.

Our bi-monthly sessions provide a platform for in-depth analysis and exchange of ideas among experts and enthusiasts in the field. ​By examining case studies, current research, and emerging trends, we aim to foster a deeper understanding of the implications of ​multi-agent AI systems on society, ethics, and governance.

Organiser & Facilitator: Dorcas Nyamwaya

Time: Every other Saturday at 7 PM, EAT (GMT+3)

Venue: Virtual

Up-Coming Meetings

2024.08.03: Liu et al., 2024, “Agent Design Pattern Catalogue: A Collection of Architectural Patterns for Foundation Model ​based Agents”

Past readings and discussions (most recent first):

• 2024.02.22: Reading Group on Break, Meeting Cancelled
• 2024.06.08: Chan et al., 2024, “Visibility into AI Agents”.
• 2024.05.25: Lu et al., 2024, “Towards Responsible Generative AI: A Reference Architecture for Designing Foundation ​Model based Agents”.
• 2024.05.11: Kolt, 2024, “Governing AI Agents”.
• 2024.04.27: Han et al., 2024, “LLM Multi-Agent Systems: Challenges and Open Problems”.
  
• 2024.04.13: Durante et al., 2024, “Agent AI: Surveying the Horizons of Multimodal Interaction”.
  
• 2024.03.30: Wang et al., 2024, “A Survey on Large Language Model based Autonomous Agents”.
  
• 2024.03.16: Shavit et al., 2023, “Practices for Governing Agentic AI Systems”.
  
• 2024.03.02: Yuan et al., 2023, “A Survey of Progress on Cooperative Multi-agent Reinforcement Learning in Open ​Environment”.
  
• 2024.02.17: Yocrum et al., 2023, “Mitigating Generative Agent Social Dilemmas”.
  
• 2024.02.03: Xi et al., 2023, “The Rise and Potential of Large Language Model Based Agents: A Survey”.
  
• 2024.01.20: Chan et al., 2023, “Harms from Increasingly Agentic Algorithmic Systems”.
  

Join us to engage in meaningful conversations and ​contribute to the development of effective policies ​for managing the risks associated with multi-agent ​AI systems

Express your interest in our programs and join our community.