Seminar - Tao Chen

Event Date: 

Wednesday, October 24, 2018 - 3:30pm

Event Location: 

  • Buchanan 1930

Title: Nonparametric Adaptive Robust Control under Model Uncertainty

Abstract:

Adaptive robust method is developed to solve discrete-time Markovian control problem subject to model uncertainty, where reduction of uncertainty through learning is taken into consideration. In this work, we extend the method to the nonparametric setup. To show the existence of measurable selectors in dynamic programming, we embed the distributions of real valued random variables into the space of probability measures on the extended real line equipped with Levy-Prokhorov metric. We also present a minimax result for the Bellman equations. Possible numerical method for solving the problem is discussed.

Bio:

Tao Chen is a Visiting Assistant Professor at the Department of Statistics and Applied Probability at UCSB.

Website