Improving Deep Learning Model Robustness Against Adversarial Attack by Increasing the Network Capacity

Marco Marchetti, Edmond S. L. Ho*

*Corresponding author for this work

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

6 Downloads (Pure)

Abstract

Nowadays, we are more and more reliant on Deep Learning (DL) models and thus it is essential to safeguard the security of these systems. This paper explores the security issues in Deep Learning and analyses, through the use of experiments, the way forward to build more resilient models. Experiments are conducted to identify the strengths and weaknesses of a new approach to improve the robustness of DL models against adversarial attacks. The results show improvements and new ideas that can be used as recommendations for researchers and practitioners to create increasingly better DL algorithms.
Original languageEnglish
Title of host publicationAdvances in Cybersecurity, Cybercrimes, and Smart Emerging Technologies
Subtitle of host publicationCCSET 2022
EditorsAhmed A. Abd El-Latif, Yassine Maleh, Wojciech Mazurczyk, Mohammed ELAffendi, Mohamed I. Alkanhal
Place of PublicationCham, Switzerland
PublisherSpringer
Pages85-96
Number of pages12
ISBN (Electronic)9783031211010
ISBN (Print)9783031211003
DOIs
Publication statusPublished - 12 Mar 2023
EventInternational conference on Cybersecurity, Cybercrimes, and Smart Emerging Technologies - Prince Sultan University, Saudi Arabia
Duration: 10 May 202211 May 2022

Publication series

NameEngineering Cyber-Physical Systems and Critical Infrastructures
PublisherSpringer
Volume4
ISSN (Print)2731-5002
ISSN (Electronic)2731-5010

Conference

ConferenceInternational conference on Cybersecurity, Cybercrimes, and Smart Emerging Technologies
Abbreviated titleCCSET2022
Country/TerritorySaudi Arabia
Period10/05/2211/05/22

Fingerprint

Dive into the research topics of 'Improving Deep Learning Model Robustness Against Adversarial Attack by Increasing the Network Capacity'. Together they form a unique fingerprint.

Cite this