Advanced Topics in Systems, Control and Learning

More Information

UBC Calendar

3 Credits

EECE 571U

Course Description

The objective of this course is to develop an in-depth understanding of the fundamental tools for control and learning and to apply this understanding in cutting-edge research.  We will discuss tools for studying dynamical systems in state-space domain including the existence and uniqueness of solutions, stability and optimal control.  We will then discuss fundamentals of online optimization for learning and control.  Building on this background, we will study recent papers in the area of safe learning and control, as well as multi-agent learning.  The course participants will apply the tools to a domain specific application of their own choice through a course project.

Topic Outline

Linear systems theory: fundamentals of linear and Hilbert space theory for characterizing solutions of dynamical systems, stability, controllability

Optimal control theory: stochastic systems, dynamic programming, approximate dynamic programming

Learning and control: online and stochastic optimization, safe learning, game theory and multi-agent learning.

Marking Scheme

in-class participation 20%, project 60%, presentation: 20%

Professor: 

a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949
Email:

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2020 The University of British Columbia