Professor Mihailo Jovanovic at ECE Spring 2023 Colloquium

Robustness of gradient methods for data-driven decision making

Gradient descent and its accelerated variants are increasingly used for learning and data-driven control of uncertain dynamical systems in which an approximation of the gradient is sought through noisy measurements. In the first part of the talk, we utilize techniques from control theory to quantify robustness of accelerated first-order algorithms to stochastic uncertainties in gradient evaluation. We identify fundamental tradeoffs between noise amplification and convergence rates for any acceleration scheme similar to Nesterov's and heavy-ball methods. We also examine performance and efficiency of model-free reinforcement learning methods which attempt to find an optimal control action for an unknown dynamical system by searching over the parameter space of controllers. For systems with unknown state-space parameters, we provide theoretical bounds on the convergence rate and sample complexity of the random search method applied to the standard linear quadratic regulator problem.

About Prof. Mihailo Jovanovic

Mihailo Jovanovic is a professor in the Ming Hsieh Department of Electrical and Computer Engineering and the founding director of the Center for Systems and Control at the University of Southern California. He was a faculty member in the Department of Electrical and Computer Engineering at the University of Minnesota, Minneapolis, from 2004 until 2017, and has held visiting positions with Stanford University, the Institute for Mathematics and its Applications, and the Simons Institute for the Theory of Computing. Prof. Jovanovic received a CAREER Award from the National Science Foundation in 2007, the George S. Axelby Outstanding Paper Award from the IEEE Control Systems Society in 2013, and the Distinguished Alumnus Award from the University of California at Santa Barbara in 2014. He is a Fellow of the American Physical Society (APS) and the Institute of Electrical and Electronics Engineers (IEEE). 

Start date
Thursday, April 27, 2023, 4 p.m.
End date
Thursday, April 27, 2023, 5 p.m.

Share