Scalable Incremental Robustness Analysis of Neural Network Feedback Systems
Abstract
Semidefinite programming (SDP) certificates for feedback systems containing deep neural networks (NNs) typically scale with the total number of neurons, whereas small-gain tests are scalable but can be highly conservative. This paper develops a unified and scalable framework for incremental robust stability and performance analysis of feedback interconnections involving high-dimensional NNs and unmodeled dynamics. By combining a structured decomposition of the full-order SDP condition with scalable Lipschitz constant estimation algorithms, we derive reduced verification conditions that certify incremental convergence and incremental ℓ2-gain bounds. The dimensions of the resulting control-analysis linear matrix inequalities (LMIs) depend only on the widths of the last two network layers and are \textit{independent of network depth}. The framework preserves the coupling between the plant and the NN, with the incremental small-gain condition recovered as a special case. To further reduce conservatism, we develop a multi-round alternating update scheme that iteratively refines the coupling variables while preserving scalability. Numerical experiments show that the proposed framework achieves state-of-the-art incremental ℓ2-gain bounds for large-scale NN feedback systems.
Related
- Enhanced Feedback Mechanisms for Resource-Efficient Incremental Redundancy
- Incremental Stability of Gradient Flows under Generalized Polyak-Łojasiewicz Inequalities and Its Implications for General Systems
- Effect of a fourth-order diffusion term on the robustness of the transfer function stability for hyperbolic systems
- GNet: A scalable and flexible Gaussian process network with nonparametric neurons
- Interpretable Intelligent Fault Diagnosis of Active Magnetic Bearing–Rotor Systems Based on Multidimensional Feature Fusions with Robustness Assessment
BibTeX
@misc{wang2026scalable,
title = {{Scalable Incremental Robustness Analysis of Neural Network Feedback Systems}},
author = {Zichen Wang and Peter Seiler and Geir Dullerud and Bin Hu},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.22576},
eprint = {2609.22576},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.22576},
}
RIS
TY - UNPB
TI - Scalable Incremental Robustness Analysis of Neural Network Feedback Systems
AU - Zichen Wang
AU - Peter Seiler
AU - Geir Dullerud
AU - Bin Hu
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.22576
UR - https://arxiv.org/abs/2609.22576
AB - Semidefinite programming (SDP) certificates for feedback systems containing deep neural networks (NNs) typically scale with the total number of neurons, whereas small-gain tests are scalable but can be highly conservative. This paper develops a unified and scalable framework for incremental robust stability and performance analysis of feedback interconnections involving high-dimensional NNs and unmodeled dynamics. By combining a structured decomposition of the full-order SDP condition with scalable Lipschitz constant estimation algorithms, we derive reduced verification conditions that certify incremental convergence and incremental ℓ_2-gain bounds. The dimensions of the resulting control-analysis linear matrix inequalities (LMIs) depend only on the widths of the last two network layers and are \textit{independent of network depth}. The framework preserves the coupling between the plant and the NN, with the incremental small-gain condition recovered as a special case. To further reduce conservatism, we develop a multi-round alternating update scheme that iteratively refines the coupling variables while preserving scalability. Numerical experiments show that the proposed framework achieves state-of-the-art incremental ℓ_2-gain bounds for large-scale NN feedback systems.
ER -