Automated Functional Decomposition for Hybrid Zonotope Over-approximations with Application to LSTM Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Glunt, Jonah J., Siefert, Jacob A., Thompson, Andrew F., Ruths, Justin, Pangborn, Herschel C.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909543260225536
author Glunt, Jonah J.
Siefert, Jacob A.
Thompson, Andrew F.
Ruths, Justin
Pangborn, Herschel C.
author_facet Glunt, Jonah J.
Siefert, Jacob A.
Thompson, Andrew F.
Ruths, Justin
Pangborn, Herschel C.
contents Functional decomposition is a powerful tool for systems analysis because it can reduce a function of arbitrary input dimensions to the sum and superposition of functions of a single variable, thereby mitigating (or potentially avoiding) the exponential scaling often associated with analyses over high-dimensional spaces. This paper presents automated methods for constructing functional decompositions used to form set-based over-approximations of nonlinear functions, with particular focus on the hybrid zonotope set representation. To demonstrate these methods, we construct a hybrid zonotope set that over-approximates the input-output graph of a long short-term memory neural network, and use functional decomposition to represent a discrete hybrid automaton via a hybrid zonotope.
format Preprint
id arxiv_https___arxiv_org_abs_2503_15336
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Automated Functional Decomposition for Hybrid Zonotope Over-approximations with Application to LSTM Networks
Glunt, Jonah J.
Siefert, Jacob A.
Thompson, Andrew F.
Ruths, Justin
Pangborn, Herschel C.
Systems and Control
Functional decomposition is a powerful tool for systems analysis because it can reduce a function of arbitrary input dimensions to the sum and superposition of functions of a single variable, thereby mitigating (or potentially avoiding) the exponential scaling often associated with analyses over high-dimensional spaces. This paper presents automated methods for constructing functional decompositions used to form set-based over-approximations of nonlinear functions, with particular focus on the hybrid zonotope set representation. To demonstrate these methods, we construct a hybrid zonotope set that over-approximates the input-output graph of a long short-term memory neural network, and use functional decomposition to represent a discrete hybrid automaton via a hybrid zonotope.
title Automated Functional Decomposition for Hybrid Zonotope Over-approximations with Application to LSTM Networks
topic Systems and Control
url https://arxiv.org/abs/2503.15336