Stability-Preserving Online Adaptation of Neural Closed-loop Maps
Danilo Saccani, Luca Furieri, Giancarlo Ferrari-Trecate · 2026
Read Stability-Preserving Online Adaptation of Neural Closed-loop Maps by Danilo Saccani, Luca Furieri, Giancarlo Ferrari-Trecate (2026) 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-2603-22469.txt.
Description
Saccani and co-authors tackle a hidden risk in updating neural controllers while a system runs. Swapping one stabilizing policy for another can itself break stability. They model each controller as an operator with bounded gain and derive rules for safe updates. Two schemes follow: updates on a fixed schedule, and updates triggered by the state. Either way, the closed loop stays stable after any number of swaps. The team shows gains over static and naive updating on nonlinear test systems.
Catalog record
| CREATOR | Danilo Saccani, Luca Furieri, Giancarlo Ferrari-Trecate |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45893 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.22469