16:00 - 17:00
online
Title: Kernel-based models for system analysis
Abstract: This talk introduces a computational framework to identify nonlinear input-output operators from data. The goal is to find operators that fit a set of system trajectories while satisfying prior knowledge in the form of integral quadratic constraints. The data fitting algorithm is thus regularized by suitable input-output properties required for system analysis and control design. This biased identification problem is shown to admit the tractable solution of a regularized least squares problem when formulated in a suitable reproducing kernel Hilbert space. The kernel-based framework is a departure from the prevailing state-space framework. It is motivated by fundamental limitations of nonlinear state-space models at combining the fitting requirements of data-based modeling with the input-output requirements of system analysis and physical modeling.
The colloquium will take place online in Google Meet. You can email the organizer for a link to the meeting.