Online Nonstochastic Control
Description & Basic Information
This tutorial focuses on the new methodologies of non-stochastic control, regret minimization in reinforcement learning and differentiable reinforcement learning.
We will start from the basics of online convex optimization theory, compare and contrast with statistical learning, and compare to other approaches in control notably optimal, robust and adaptive control.
The literature in the "References" section covers the most recent research results in this young field, that are the focus of the tutorial.
For more classical texts see "extra reading" at the bottom of the References page, as well as the lecture notes.