Neural Identification of Feedback-Stabilized Nonlinear Systems

Mahrokh G. Boroujeni, Laura Meroi, Leonardo Massai, Clara L. Galimberti, and Giancarlo Ferrari-Trecate · 2025

Read Neural Identification of Feedback-Stabilized Nonlinear Systems by Mahrokh G. Boroujeni, Laura Meroi, Leonardo Massai, Clara L. Galimberti, and Giancarlo Ferrari-Trecate (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-22601.txt.

Description

Mahrokh Boroujeni and co-authors tackle, in a 2025 paper, learning a model of a nonlinear system from data gathered while a controller was running it. Feedback during collection tangles the data, which trips up standard methods. A useful model should also be stabilized by that same controller.

Older methods lean on exact noise models, wobble when data changes, or cost too much to compute. A trick called the dual Youla parameterization solves these issues for linear systems, with guarantees. Extending it to nonlinear systems adds real complexity, the authors note. Their work carries that extension forward using neural networks. Better closed-loop models matter wherever machines are tuned from their own operating logs.

Catalog record

CREATORMahrokh G. Boroujeni, Laura Meroi, Leonardo Massai, Clara L. Galimberti, and Giancarlo Ferrari-Trecate
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-22601

Browse the full library · About the Tower of Babel