HyParLyVe: Hyperplane Partitioning for Neural Lyapunov Verification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wayment, Jesse, Yarbrough, Brian, Wang, Jingbo, Sundaram, Shreyas, Paré, Philip E.
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915981529448448
author Wayment, Jesse
Yarbrough, Brian
Wang, Jingbo
Sundaram, Shreyas
Paré, Philip E.
author_facet Wayment, Jesse
Yarbrough, Brian
Wang, Jingbo
Sundaram, Shreyas
Paré, Philip E.
contents This work introduces HyParLyVe (Hyperplane Partitioned Lyapunov Verifier), a novel algorithm for sound and complete verification of neural Lyapunov candidates by interpreting shallow ReLU networks as hyperplane arrangements. This perspective reduces positive definiteness verification to a finite set of vertex evaluations, and the decrease condition to a bounded optimization problem over each region. We formally prove correctness of the proposed verification procedures and demonstrate that HyParLyVe achieves significant speedups over state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2605_03992
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle HyParLyVe: Hyperplane Partitioning for Neural Lyapunov Verification
Wayment, Jesse
Yarbrough, Brian
Wang, Jingbo
Sundaram, Shreyas
Paré, Philip E.
Systems and Control
This work introduces HyParLyVe (Hyperplane Partitioned Lyapunov Verifier), a novel algorithm for sound and complete verification of neural Lyapunov candidates by interpreting shallow ReLU networks as hyperplane arrangements. This perspective reduces positive definiteness verification to a finite set of vertex evaluations, and the decrease condition to a bounded optimization problem over each region. We formally prove correctness of the proposed verification procedures and demonstrate that HyParLyVe achieves significant speedups over state-of-the-art methods.
title HyParLyVe: Hyperplane Partitioning for Neural Lyapunov Verification
topic Systems and Control
url https://arxiv.org/abs/2605.03992