Chelsea Sidrane joins MPI-IS as new group leader

Tübingen – Chelsea Sidrane was awarded an Emmy Noether Fellowship by the German Research Foundation DFG to build an independent research group named “Verification And Learning for Intelligent Systems”.

Sidrane’s research is focused on reliability in machine learning for robotics. Machine learning is used everywhere these days, but when programmers train complex neural networks, they cannot easily analyze what is going on inside. Unfortunately, research has shown that these models contain bugs known as adversarial examples. This poses a safety risk if one is going to use neural policies on robot hardware in the real world. Sidrane’s work uses techniques from optimization and formal methods to verify precise mathematical properties of neural networks used as subcomponents in dynamical systems.

Chelsea Sidrane is currently recruiting Ph.D. students and research assistants. One can apply to work with her through four Ph.D. programs: IMPRS-IS, CLS, ELLIS, and MP-AIX with the first deadlines at the end of October. Or find her email here: https://www.chelseasidrane.com

Previously, Chelsea Sidrane was a Digital Futures Postdoctoral Fellow at the KTH Royal Institute of Technology in Stockholm where she worked with Jana Tumova in the Divison of Robotics, Perception and Learning (RPL). Sidrane is a Ph.D. graduate from the Stanford Intelligent Systems Lab (SISL) in the Stanford University Department of Aeronautics and Astronautics, where she was advised by Mykel Kochenderfer.