Course positioning
MS&E 318 is a PhD-level seminar on making AI systems not only powerful but provably safe. It covers classical constrained optimization, safe RL (trust regions, Lyapunov, reachability, shielding), hybrid ML/optimization (projection networks, solver-in-the-loop), and LLM alignment, with applications in robotics, finance, healthcare, and energy.
What it covers
- Constrained optimization: Lagrangians, robust/stochastic programming
- Safe RL: trust regions, Lyapunov, reachability, shielding
- Hybrid methods: projection networks, solver-in-the-loop
- LLM alignment: fine-tuning, model editing, tool use, interactive alignment
- Theoretical results: feasibility, stability, guarantees
- Applications and project
Prerequisites
Optimization at the CME 307 or EE 364A level.
Assessment & workload
Paper discussion plus a final project. Check the current syllabus for details.
Who it's for
Graduate students in safe RL, control, optimization, or alignment; people who want AI with guarantees.
Study advice
Brush up on convex optimization and control theory first. The course is theoretical; pick a project where safety can be proven or measured.
Related courses
Official course description
How can we design AI systems that are not only powerful but also provably safe and trustworthy? This advanced PhD seminar surveys algorithmic methods to enforce hard constraints in machine learning, reinforcement learning, and generative AI. Topics include classical constrained optimization (Lagrangian methods, robust and stochastic programming), safe reinforcement learning (trust regions, Lyapunov functions, reachability, shielding), hybrid ML-optimization methods (projection networks, solver-in-the-loop architectures), and alignment strategies for large language models (fine-tuning, model editing, tool use, and interactive alignment). Each week highlights a key theoretical result alongside state-of-the-art research, with applications spanning robotics, finance, healthcare, energy, and language models. Students will critically assess the strengths and limitations of these methods and develop final projects that apply or extend them in real-world domains. Prerequisites: optimization at the level of CME 307 or EE 364a.