Nathan Fulton, Ph.D.
Research Staff Member
Lots of exciting news this month! Nathan Hunt has a paper on end-to-end RL that will appear at HSCC, Kody Vajjha has a paper on a Coq formalization of convergence of value and policy iteration will be presented at this year's CPP, and I am serving on the PC for this year's Conference on Automated Deduction.
Nathan was a core developer of the KeYmaera X theorem prover for hybrid systems, which he developed together with other members of the Logical Systems Lab while earning his Ph.D. in Carnegie Mellon University's Computer Science Department. KeYmaera X demonstrates how formal methods techniques can improve the safety and trustworthiness of autonomous planes, magnetic control systems, and embedded devices. Nathan also uses hybrid systems verification tooling to build verifiable safety guarantees for reinforcement learning algorithms. He also collaborates with researchers at MIT and IBM on theory and tooling for other types of neural-symbolic systems.
Nathan maintains an active interest in undergraduate and high school education and mentoring. He is married to Abby Pekoske, who is currently a Ph.D. student working on DNA modeling with David Swigon in the University of Pittsburgh's Coplex Biological Systems Group.