Integral Quadratic Constraints for Repeated ReLU

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Noori, Sahel Vahedi, Hu, Bin, Dullerud, Geir, Seiler, Peter
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912972215943168
author Noori, Sahel Vahedi
Hu, Bin
Dullerud, Geir
Seiler, Peter
author_facet Noori, Sahel Vahedi
Hu, Bin
Dullerud, Geir
Seiler, Peter
contents This paper presents a new dynamic integral quadratic constraint (IQC) for the repeated Rectified Linear Unit (ReLU). These dynamic IQCs can be used to analyze stability and induced $\ell_2$-gain performance of discrete-time, recurrent neural networks (RNNs) with ReLU activation functions. These analysis conditions can be incorporated into learning-based controller synthesis methods, which currently rely on static IQCs. We show that our proposed dynamic IQCs for repeated ReLU form a superset of the dynamic IQCs for repeated, slope-restricted nonlinearities. We also prove that the $\ell_2$-gain bounds are nonincreasing with respect to the horizon used in the dynamic IQC filter. A numerical example using a simple (academic) RNN shows that our proposed IQCs lead to less conservative bounds than existing IQCs.
format Preprint
id arxiv_https___arxiv_org_abs_2603_17137
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Integral Quadratic Constraints for Repeated ReLU
Noori, Sahel Vahedi
Hu, Bin
Dullerud, Geir
Seiler, Peter
Systems and Control
This paper presents a new dynamic integral quadratic constraint (IQC) for the repeated Rectified Linear Unit (ReLU). These dynamic IQCs can be used to analyze stability and induced $\ell_2$-gain performance of discrete-time, recurrent neural networks (RNNs) with ReLU activation functions. These analysis conditions can be incorporated into learning-based controller synthesis methods, which currently rely on static IQCs. We show that our proposed dynamic IQCs for repeated ReLU form a superset of the dynamic IQCs for repeated, slope-restricted nonlinearities. We also prove that the $\ell_2$-gain bounds are nonincreasing with respect to the horizon used in the dynamic IQC filter. A numerical example using a simple (academic) RNN shows that our proposed IQCs lead to less conservative bounds than existing IQCs.
title Integral Quadratic Constraints for Repeated ReLU
topic Systems and Control
url https://arxiv.org/abs/2603.17137