PhD Positions in Reinforcement Learning at RIT: Generative Control and Safety
Ali Baheri
We are seeking 1-2 exceptional PhD candidates to join our research program in Reinforcement Learning (RL) at the Rochester Institute of Technology (RIT). These positions focus on developing theoretical foundations and algorithmic frameworks at the intersection of Control Theory, Reinforcement Learning, and Generative Modeling (specifically Flow Matching).
Research Focus
The successful candidates will investigate fundamental questions in Safe and High-Performance RL, focusing on:
Generative Control via Flow Matching: Developing control-theoretic and RL algorithms using Generative Models (e.g., Flow Matching) to parameterize or synthesize optimal/safe policies and state trajectories.
Verifiable Safety Guarantees: Creating theoretical frameworks for providing verifiable, real-time safety guarantees for learning-based control systems.
Constraint Satisfaction: Investigating methods for incorporating complex constraints into sequential decision-making using generative policy representations.
Background and Qualifications
Required:
Master's degree in Control Theory, Computer Science, Applied Mathematics, or a related quantitative field.
Strong mathematical background in Optimization, Control Theory (especially nonlinear/constrained control), and Machine Learning/Reinforcement Learning.
Expertise with Generative Models (e.g., Flow Matching, Diffusion Models) is highly desirable.
How to Apply
To apply for this position, please send your CV, a brief statement of research interests, and copies of any relevant publications (if available) to:
Best regards,
Ali Baheri