Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/137946
Title: Demystifying adversarial attacks on neural networks
Authors: Yip, Lionell En Zhi
Keywords: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
Issue Date: 2020
Publisher: Nanyang Technological University
Project: SCSE19-0306
Abstract: Prevalent use of Neural Networks for Classification Tasks has brought to attention the security and integrity of the Neural Networks that industries are so reliant on. Adversarial examples are conspicuous to humans, but neural networks struggle to correctly classify images with the presence of adversarial perturbations. I introduce a framework for understanding how neural networks perceive inputs, and its relation to adversarial attack methods. I demonstrate that there is no correlation between the region of importance and the region of attack. I demonstrate that a frequently perturbed region of an adversarial example across a class in a data-set exists. I attempt to improve classification performance by exploiting the differences of input and adversarial attack, and I demonstrate a novel augmentation method for improving prediction performance of adversarial samples.
URI: https://hdl.handle.net/10356/137946
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
NTU_SCSE19-0306-U1721954J.pdf
  Restricted Access
2.4 MBAdobe PDFView/Open

Page view(s)

220
Updated on Nov 30, 2022

Download(s) 50

49
Updated on Nov 30, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.