An Abstraction-Based Framework for Neural Network Verification

Yizhak Yisrael Elboher, Justin Gottschlich, Guy Katz*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

60 Scopus citations

Abstract

Deep neural networks are increasingly being used as controllers for safety-critical systems. Because neural networks are opaque, certifying their correctness is a significant challenge. To address this issue, several neural network verification approaches have recently been proposed. However, these approaches afford limited scalability, and applying them to large networks can be challenging. In this paper, we propose a framework that can enhance neural network verification techniques by using over-approximation to reduce the size of the network—thus making it more amenable to verification. We perform the approximation such that if the property holds for the smaller (abstract) network, it holds for the original as well. The over-approximation may be too coarse, in which case the underlying verification tool might return a spurious counterexample. Under such conditions, we perform counterexample-guided refinement to adjust the approximation, and then repeat the process. Our approach is orthogonal to, and can be integrated with, many existing verification techniques. For evaluation purposes, we integrate it with the recently proposed Marabou framework, and observe a significant improvement in Marabou’s performance. Our experiments demonstrate the great potential of our approach for verifying larger neural networks.

Original languageAmerican English
Title of host publicationComputer Aided Verification - 32nd International Conference, CAV 2020, Proceedings
EditorsShuvendu K. Lahiri, Chao Wang
PublisherSpringer
Pages43-65
Number of pages23
ISBN (Print)9783030532871
DOIs
StatePublished - 2020
Event32nd International Conference on Computer Aided Verification, CAV 2020 - Los Angeles, United States
Duration: 21 Jul 202024 Jul 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12224 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference32nd International Conference on Computer Aided Verification, CAV 2020
Country/TerritoryUnited States
CityLos Angeles
Period21/07/2024/07/20

Bibliographical note

Publisher Copyright:
© 2020, The Author(s).

Fingerprint

Dive into the research topics of 'An Abstraction-Based Framework for Neural Network Verification'. Together they form a unique fingerprint.

Cite this