The statistical foundations of learning to control
Given the dramatic successes in machine learning over the past half decade, there has been a resurgence of interest in applying learning techniques to continuous control problems in robotics, self-driving cars, and unmanned aerial vehicles. Though such applications appear to be straightforward generalizations of what is known as reinforcement learning, few fundamental baselines have been established prescribing how well one must know a system in order to control it. In this talk, I will discuss how one might merge techniques from statistical learning theory with robust control to derive baselines for such continuous control. I will explore several examples that balance parameter identification against controller design and demonstrate finite sample tradeoffs between estimation fidelity and desired control performance. I will describe how these simple baselines give us insights into shortcomings of existing reinforcement learning methodology. I will close by listing several exciting open problems that must be solved before we can build robust, safe learning systems that interact with an uncertain physical environment.
Date:
13 April 2018, 14:00
Venue:
Information Engineering, Banbury Road OX1 3PH
Venue Details:
Lecture Room 7, Information Engineering Building, Engineering Science, University of Oxford
Speaker:
Professor Benjamin Recht (University of California at Berkeley)
Organising department:
Department of Engineering Science
Organisers:
Professor Kostas Margellos (University of Oxford),
Professor Antonis Papachristodoulou (University of Oxford)
Organiser contact email address:
antonis@eng.ox.ac.uk
Host:
Professor Antonis Papachristodoulou (University of Oxford)
Part of:
Control Seminar Series
Topics:
Booking required?:
Not required
Audience:
Members of the University only
Editor:
Antonis Papachristodoulou
