Enforcing Robust Control Guarantees within Neural Network Policies

Priya Donti - Carnegie Mellon University

Nov. 19, 2021, 2:30 p.m. - Nov. 19, 2021, 3:30 p.m.

Virtual (see link below)

Hosted by: David Rolnick


Abstract: When designing controllers for safety-critical systems, practitioners often face a challenging tradeoff between robustness and performance. While robust control methods provide rigorous guarantees on system stability under certain worst-case disturbances, they often result in simple controllers that perform poorly in the average (non-worst) case. In contrast, nonlinear control methods trained using deep learning have achieved state-of-the-art performance on many control tasks, but often lack robustness guarantees. In this talk, I will propose a technique that combines the strengths of these two approaches: a generic nonlinear control policy class, parameterized by neural networks, that nonetheless enforces the same provable robustness criteria as robust control. Specifically, our approach entails integrating custom convex-optimization-based projection layers into a neural network-based policy. I will demonstrate the power of this approach on several domains, and show that it improves performance over existing robust control methods while improving stability over (non-robust) reinforcement learning methods.

Bio: Priya Donti is a Ph.D. student in Computer Science and Public Policy at Carnegie Mellon University. She is also a co-founder and chair of Climate Change AI, an initiative to catalyze impactful work in climate change and machine learning. Her work focuses on machine learning for forecasting, optimization, and control in high-renewables power grids. Specifically, her research explores methods to incorporate the physics and hard constraints associated with electric power systems into deep learning models. Priya is a recipient of the MIT Technology Review "35 Innovators Under 35" award, the Siebel Scholarship, the U.S. Department of Energy Computational Science Graduate Fellowship, and best paper awards at ICML (honorable mention), ACM e-Energy (runner-up), PECI, the Duke Energy Data Analytics Symposium, and the NeurIPS workshop on AI for Social Good.

Zoom link: https://mcgill.zoom.us/j/84280715850

Virtual reception after the talk in Gather: https://gather.town/app/tYHHMh7tPcPw9037/reception