|Date: Wednesday, September 23, 2020
Location: https://umich.zoom.us/j/95407665241 Virtual (4:00 PM to 5:00 PM)
Title: Adaptive Robust Stochastic Control
Abstract: In this talk, we will review the adaptive robust control methodology that we developed for dealing with (Markovian) stochastic control problems under model uncertainty. To address the uncertainty of the model parameters for the underlying stochastic process, this novel robust technique reduces the uncertainty through online learning and updating of the parameter set while optimizing. We will also discuss the difficulties that arise in the implementation of such approach and propose a new machine learning technique to recast the numerical implementation into a statistical learning problem. Finally, we will use several examples to show that the adaptive robust approach can outperform other traditional/popular robust techniques.
Speaker: Tao Chen