Toward Safe and Scalable Multi-Agent Systems: Bridging Control, RL, and Generative AI - ACC 2026 Workshop

Organizers

Runyu (Cathy) Zhang (MIT)

Gioele Zardini (MIT)

Na Li (Harvard)

Relevant Info and Links

• When? On May 26th, 2026.
• Where? At ACC 2026, Room: TBD
• IEEE ACC 2026 Webpage

Workshop Abstract

Modern engineering and AI systems increasingly rely on large networks of interacting agents, from autonomous vehicles and robotic swarms to infrastructure and communication networks. Yet, coordinating these systems safely and efficiently at scale remains one of the hardest open problems in control and learning. As data-driven methods and foundation models pervade decision-making, the need for principled approaches that bridge learning, control, and safety has never been more pressing.

This workshop confronts this challenge head-on. It brings together researchers from control theory, reinforcement learning, and generative AI to rethink how multi-agent systems can learn, adapt, and coordinate reliably. Topics include safety-critical reinforcement learning, scalable coordination under uncertainty, and the use of generative models as priors or planners for distributed decision-making, all aimed at pushing the boundary between theoretical guarantees and practical scalability.

Speakers

Bassam Bamieh (UCSB)

Anastasia Bizyaeva (Cornell)

Ceyhun Eksin (Taxes A&M)

Andreas Malikopoulos (Cornell)

Nader Motee (Lehigh)

Dimitra Panagou (University of Michigan)

Tentative Program

The full-day (8:50am -- 5pm) workshop program will feature a series of invited talks, each followed by dedicated time for questions and discussion (25-minute presentations and 5-minute Q&A). The workshop will include a morning and an afternoon session, concluding with a panel discussion.