Reluplex made more practical: Leaky ReLU

  • Jin Xu
  • , Zishan Li
  • , Bowen Du
  • , Miaomiao Zhang*
  • , Jing Liu*
  • *Corresponding author for this work

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

408 Scopus citations

Abstract

In recent years, Deep Neural Networks (DNNs) have been experiencing rapid development and have been widely used in various fields. However, while DNNs have shown strong capabilities, their security problems have gradually been exposed. Therefore, the formal guarantee of neural network output is needed. Prior to the appearance of the Reluplex algorithm, the verification of DNNs was always a difficult problem. Reluplex algorithm is specially used to verify DNNs with ReLU activation function. This is an excellent and effective algorithm, but it cannot verify more activation functions. ReLU activation function will bring about "Dead Neuron"problem, and Leaky ReLU activation function can solve this problem, so it is necessary to verify DNNs based on Leaky ReLU activation function. Therefore, we propose the Leaky-Reluplex algorithm, which is based on the Reluplex algorithm. Leaky-Reluplex algorithm can verify DNNs based on Leaky ReLU activation function.

Original languageEnglish
Title of host publication2020 IEEE Symposium on Computers and Communications, ISCC 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728180861
DOIs
StatePublished - Jul 2020
Event2020 IEEE Symposium on Computers and Communications, ISCC 2020 - Rennes, France
Duration: 7 Jul 202010 Jul 2020

Publication series

NameProceedings - IEEE Symposium on Computers and Communications
Volume2020-July
ISSN (Print)1530-1346

Conference

Conference2020 IEEE Symposium on Computers and Communications, ISCC 2020
Country/TerritoryFrance
CityRennes
Period7/07/2010/07/20

Keywords

  • Deep Neural Networks
  • Leaky ReLU
  • Reluplex Algorithm

Fingerprint

Dive into the research topics of 'Reluplex made more practical: Leaky ReLU'. Together they form a unique fingerprint.

Cite this