Machine Learning Seminar Series Fall 2026 | Machine Learning with Hard Constraints

Abstract: How can we harness the expressive power of modern machine learning while ensuring that its outputs respect non-negotiable physical, safety, and operational constraints? Penalizing violations during training encourages compliance but does not guarantee it at deployment. In this talk, I will present a principled approach to incorporating hard constraints into both learning architectures and generative sampling. I will begin with hard-constrained neural networks (HardNets), which satisfy input-dependent constraints by construction while retaining universal approximation of feasible functions. I will then show how these ideas enable various applications, from learned models of chaotic dynamics with provably bounded trajectories through energy-constrained operator learning to scalable safe reinforcement learning and learning-based control with formal guarantees. Finally, I will turn to generative models, namely flow-matching and diffusion models, with hard inference-time constraints. By recasting constrained generation as a trajectory optimization problem, we can use receding-horizon control to steer pretrained models toward feasible outputs without unnecessarily restricting their sampling dynamics or requiring retraining. Applications spanning fluid dynamics, robotic control and planning, PDE control, and language-guided image editing illustrate a unifying principle: hard constraints should define what a model is allowed to do without unnecessarily limiting what it can learn or generate.

Bio: Navid Azizan is the Alfred H. (1929) and Jean M. Hayes Associate Professor at MIT, where he holds dual appointments in the Department of Mechanical Engineering (in Control, Instrumentation & Robotics) and the Schwarzman College of Computing's Institute for Data, Systems & Society (IDSS), and is a Principal Investigator in the Laboratory for Information & Decision Systems (LIDS). His research interests broadly lie in machine learning, systems and control, and mathematical optimization. His research lab focuses on various aspects of reliable AI systems, with applications to high-stakes and safety-critical settings. He obtained his PhD in Computing and Mathematical Sciences (CMS) from the California Institute of Technology (Caltech) in 2020, his MSc in electrical engineering from the University of Southern California in 2015, and his BSc in electrical engineering and physics from Sharif University of Technology in 2013. Prior to joining MIT, he completed a postdoc at Stanford University in 2021. Additionally, he was a research scientist intern at Google DeepMind in 2019. He is a recipient of several awards, including the National Science Foundation CAREER Award, research awards from Amazon, Google, and MathWorks, among others, and the inaugural Information Theory and Applications (ITA) “Sun” (Gold) Graduation Award. His work has been recognized with best paper awards at several venues, including Learning for Dynamics and Control (L4DC), the INFORMS Junior Faculty Paper Competition, and ACM Greenmetrics. He was named in the list of Outstanding Academic Leaders in Data by the CDO Magazine for two consecutive years in 2024 and 2023. His teaching and mentorship have been recognized with the Joseph A. Martore (1975) Excellence in Teaching Award, the Frank E. Perkins Award for Excellence in Graduate Advising (MIT Institute Award), and the UROP Outstanding Mentor Award.

While we prefer you join in person, if you are unable to attend, join online at https://gatech.zoom.us/j/91508317626?pwd=WrbKsZaW2tcby2uQdVcAjsYaX0FL8l.1

Meeting ID: 915 0831 7626
Passcode: 647326

For CODA guest access, please contact shatcher8@gatech.edu at least 2 business days prior to the event.