Repository logo
 

DYNAMIC DEFENSES AND THE TRANSFERABILITY OF ADVERSARIAL EXAMPLES

dc.access.optionOpen Access
dc.contributor.advisorTabrizi, M. H. N
dc.contributor.authorThomas, Sam
dc.contributor.departmentComputer Science
dc.date.accessioned2019-06-12T20:06:51Z
dc.date.available2019-06-12T20:06:51Z
dc.date.created2019-05
dc.date.issued2019-05-02
dc.date.submittedMay 2019
dc.date.updated2019-06-11T16:00:40Z
dc.degree.departmentComputer Science
dc.degree.disciplineMS-Software Engineering
dc.degree.grantorEast Carolina University
dc.degree.levelMasters
dc.degree.nameM.S.
dc.description.abstractAdversarial machine learning has been an important area of study for the securing of machine learning systems. However, for every defense that is made to protect these artificial learners, a more sophisticated attack emerges to defeat it. This has created an arms race, with the problem of adversarial attacks never being fully mitigated. This thesis examines the field of adversarial machine learning; specifically, the property of transferability, and the use of dynamic defenses as a solution to attacks which leverage it. We show that this is an emerging field of research, which may be the solution to one of the most intractable problems in adversarial machine learning. We go on to implement a minimal experiment, demonstrating that research within this area is easily accessible. Finally, we address some of the hurdles to overcome in order to unify the disparate aspects of current related research.
dc.format.mimetypeapplication/pdf
dc.identifier.urihttp://hdl.handle.net/10342/7284
dc.language.isoen
dc.publisherEast Carolina University
dc.subjectadversarial machine learning
dc.subjecttransferability
dc.subject.lcshMachine learning
dc.subject.lcshDYNAMO (Computer program language)
dc.subject.lcshComputer security
dc.titleDYNAMIC DEFENSES AND THE TRANSFERABILITY OF ADVERSARIAL EXAMPLES
dc.typeMaster's Thesis
dc.type.materialtext

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
THOMAS-MASTERSTHESIS-2019.pdf
Size:
792.74 KB
Format:
Adobe Portable Document Format