Simplifying Neural Networks Using Formal Verification

Sumathi Gokulanathan, Alexander Feldsher, Adi Malca, Clark Barrett, Guy Katz*

*Corresponding author for this work

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

22 Scopus citations


Deep neural network (DNN) verification is an emerging field, with diverse verification engines quickly becoming available. Demonstrating the effectiveness of these engines on real-world DNNs is an important step towards their wider adoption. We present a tool that can leverage existing verification engines in performing a novel application: neural network simplification, through the reduction of the size of a DNN without harming its accuracy. We report on the work-flow of the simplification process, and demonstrate its potential significance and applicability on a family of real-world DNNs for aircraft collision avoidance, whose sizes we were able to reduce by as much as 10%.

Original languageAmerican English
Title of host publicationNASA Formal Methods - 12th International Symposium, NFM 2020, Proceedings
EditorsRitchie Lee, Susmit Jha, Anastasia Mavridou
Number of pages9
ISBN (Print)9783030557539
StatePublished - 2020
Event12th International Symposium on NASA Formal Methods, NFM 2020 - Moffett Field, United States
Duration: 11 May 202015 May 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12229 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference12th International Symposium on NASA Formal Methods, NFM 2020
Country/TerritoryUnited States
CityMoffett Field

Bibliographical note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.


  • Deep neural networks
  • Marabou
  • Simplification
  • Verification


Dive into the research topics of 'Simplifying Neural Networks Using Formal Verification'. Together they form a unique fingerprint.

Cite this