Local Stability and Region of Attraction Analysis for Neural Network Feedback Systems under Positivity Constraints
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916987530117120 |
|---|---|
| author | Hedesh, Hamidreza Montazeri Wafi, Moh Kamalul Siami, Milad |
| author_facet | Hedesh, Hamidreza Montazeri Wafi, Moh Kamalul Siami, Milad |
| contents | We study the local stability of nonlinear systems in the Lur'e form with static nonlinear feedback realized by feedforward neural networks (FFNNs). By leveraging positivity system constraints, we employ a localized variant of the Aizerman conjecture, which provides sufficient conditions for exponential stability of trajectories confined to a compact set. Using this foundation, we develop two distinct methods for estimating the Region of Attraction (ROA): (i) a less conservative Lyapunov-based approach that constructs invariant sublevel sets of a quadratic function satisfying a linear matrix inequality (LMI), and (ii) a novel technique for computing tight local sector bounds for FFNNs via layer-wise propagation of linear relaxations. These bounds are integrated into the localized Aizerman framework to certify local exponential stability. Numerical results demonstrate substantial improvements over existing integral quadratic constraint-based approaches in both ROA size and scalability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_22889 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Local Stability and Region of Attraction Analysis for Neural Network Feedback Systems under Positivity Constraints Hedesh, Hamidreza Montazeri Wafi, Moh Kamalul Siami, Milad Systems and Control Artificial Intelligence 93D09, 93D20, 93C10, 68T07 B.1.3; G.1; I.2; I.2.3; I.2.8; I.2.1; J.2 We study the local stability of nonlinear systems in the Lur'e form with static nonlinear feedback realized by feedforward neural networks (FFNNs). By leveraging positivity system constraints, we employ a localized variant of the Aizerman conjecture, which provides sufficient conditions for exponential stability of trajectories confined to a compact set. Using this foundation, we develop two distinct methods for estimating the Region of Attraction (ROA): (i) a less conservative Lyapunov-based approach that constructs invariant sublevel sets of a quadratic function satisfying a linear matrix inequality (LMI), and (ii) a novel technique for computing tight local sector bounds for FFNNs via layer-wise propagation of linear relaxations. These bounds are integrated into the localized Aizerman framework to certify local exponential stability. Numerical results demonstrate substantial improvements over existing integral quadratic constraint-based approaches in both ROA size and scalability. |
| title | Local Stability and Region of Attraction Analysis for Neural Network Feedback Systems under Positivity Constraints |
| topic | Systems and Control Artificial Intelligence 93D09, 93D20, 93C10, 68T07 B.1.3; G.1; I.2; I.2.3; I.2.8; I.2.1; J.2 |
| url | https://arxiv.org/abs/2505.22889 |