Neprijateljski napadi na modele za klasifikaciju slike

master's thesis

• Title (english): Adversarial attacks on image classification models
• Author: Krešimir Vukić
• Mentor: Siniša Šegvić (mentor)
• Committee member: Siniša Šegvić (predsjednik povjerenstva)
• Committee member: Stjepan Groš (član povjerenstva)
• Committee member: Marko Čupić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Nadzirani diskriminativni modeli sastavni su dio mnogih sustava računalnog vida. No, poznato je da su ranjivi neprijateljskim primjerima što je neprihvatljivo za osjetljive produkcijske sustave. Upravo iz toga proizlazi važnost poznavanja njihove funkcionalnosti, razloga postojanja i mogućih obrana. U ovoj tezi opisana je nekolicina postojećh neprijateljskih napada i primjenjivih obrana. Napadi FGSM i PGD te obrana treniranja sa neprijateljskim primjerima su implementirani i testirani na MNIST i SVHN podatkovnim skupovima. Također, istražen je utjecaj različitih perturbacijskih normi na utjecaj klasifikacije.

Abstract (english)

Strongly supervised discriminative models are the principal ingredient in many computer vision applications. However, discriminative models are known for being vulnerable to adversarial examples which is unacceptable for production systems. Hence, the techniques for constructing adversarial examples and building defenses against them are a very important research topic. This thesis consists of brief descriptions of existing adversarial attacks and applicable defenses. FGSM and PGD attacks, as well as defense through adversarial training, are developed and tested on the MNIST and SVHN data sets. Also, the influence of the perturbation norm on the efficiency of the attacks is briefly described.

• Keywords: neprijateljski napadi; obrana neuronske mreže; nadzirano učenje; napad pgd; napad fgsm; treniranje neprijateljskim primjerima; norma udaljenosti
• Keywords (english): adversarial attacks; neural network defenses; supervised learning; pgd attack; fgsm attack; adversarial training; perturbation norm
• Language: english
• URN:NBN: urn:nbn:hr:168:519031
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 23:02:58