I am Sunbochen Tang (汤孙博忱), a Ph.D. candidate in the AeroAstro department at the Massachusetts Institute of Technology (MIT), where I am a member of the Laboratory for Information and Decision Systems (LIDS). I am fortunate to be advised by Prof. Navid Azizan, and I have also had the privilege of collaborating with Dr. Anuradha Annaswamy at MIT and Prof. Ilya Kolmanovsky at the University of Michigan.

News

• 🏆 Our paper “ECO: Energy-Constrained Operator Learning for Chaotic Dynamics with Boundedness Guarantees” received the Best Paper Award at L4DC 2026!

• Our paper “Learning Dissipative Chaotic Dynamics with Stability Guarantees” has been accepted to the Proceedings of the National Academy of Sciences (PNAS)!

• Our paper “Reverse Flow Matching: A Unified Framework for Online Reinforcement Learning with Diffusion and Flow Policies” was accepted to ICML 2026 as a Spotlight!

Research

My research lies at the intersection of control theory, machine learning, and robotics, with a goal of developing reliable, theoretically-grounded, and efficient AI models for safety-critical and physical systems. My current work focuses on creating trustworthy learning frameworks for practical applications, with projects including:

• Safety-Constrained Guidance and Reward Fine-Tuning for Vision-Language-Action (VLA) Policies.

• Learning Chaotic Dynamics with Stability and Boundedness Guarantees.

• Learning-Based Adaptive Control for Quadrotor Agile Flight under Unknown Disturbances.

I’m passionate about these challenges and welcome conversations with others in the field. If you’d like to connect, please don’t hesitate to get in touch.

When I’m not in the lab, my interests range from outdoor activities like hiking in the White Mountains and sailing on the Charles River, to hobbies like photography and playing the piano. I also enjoy traveling and visiting museums.