Discrete State System Identification: Examples and Bounds - Professor Carolyn Beck

Discrete State System Identification: Examples and Bounds

2025 UTS FEIT Research Excellence Lecture Series - Guest Lecture

Abstract

We explore data-driven methods for modelling discrete-valued dynamical systems that evolve over networks. Examples of such systems include the spread of viruses and diseases, the propagation of ideas and misinformation, the fluctuation of stock prices, and the correlations of financial risk between banking and economic institutions. While data for these systems is often widely available, methods to identify relevant mathematical models, including the underlying network topology, are not well established or universally agreed upon. Traditional system identification methods focus on continuous-valued dynamical systems, with an emphasis on asymptotic properties, such as consistency. More recent approaches have shifted toward sample complexity, which considers how much data is needed to achieve an acceptable model approximation. In this talk, we will address the challenge of identifying a mathematical model from data for a discrete-valued, discrete-time dynamical system evolving over a network. Specifically, we will demonstrate guaranteed consistency conditions and sample complexity bounds using maximum likelihood estimation approaches. We will also discuss applications to the examples.

Speaker Biography

Carolyn, received her PhD from Caltech, her MS from Carnegie Mellon, and her BS from California State Polytechnic University, all in Electrical Engineering. Prior to her PhD studies, she worked as a Research and Development Engineer for Hewlett-Packard in Silicon Valley. She is currently a Professor at the University of Illinois at Urbana-Champaign in Industrial and Systems Engineering, and has held visiting positions at KTH (Stockholm, Sweden), Stanford University and Lund University (Sweden). She serves as an Associate Editor for the IEEE Transactions on Control of Network Systems, on the IEEE Board of Governors for the Control Systems Society (CSS) and is the President of CSS. Carolyn is an IEEE Fellow, and was the recipient of a NSF CAREER Award, an ONR Young Investigator Award, and local teaching honours. Her research interests lie in the development of model approximation methods, network inference and aggregation, and distributed optimization and control, with applications to epidemic processes and energy networks.