Interested in Learning for Safety-Critical Control?

Check talk at the "Physics ∩ ML" Seminar
http://www.physicsmeetsml.org/

Title: Learning for Safety-Critical Control in Dynamical Systems

Abstract:
This talk describes ongoing research at Caltech on integrating learning into the design of safety-critical controllers for dynamical systems. To achieve control-theoretic guarantees while using powerful function classes such as neural networks, we must carefully integrate conventional control principles with learning into unified frameworks. I will present two paradigms: integration in dynamics modeling and in policy/controller design. Special emphasis on methods that both admit relevant safety guarantees and are practical to deploy.

Featuring work by Caltech students/postdocs/alumni:
Hoang Le (http://hoangle.info/)
Guanya Shi (http://gshi.me/)
Anqi Liu (https://anqiliu-ai.github.io/)
Richard Cheng (https://rcheng805.github.io/)
Jialin Song (https://jialin-song.github.