MultiMagNet: A Non-deterministic Approach based on the Formation of Ensembles for Defending Against Adversarial Images

Gabriel Machado, Ronaldo Goldschmidt, Eugênio Silva

2019

Abstract

Deep Neural Networks have been increasingly used in decision support systems, mainly because they are the state-of-the-art algorithms for solving challenging tasks, such as image recognition and classification. However, recent studies have shown these learning models are vulnerable to adversarial attacks, i.e. attacks conducted with images maliciously modified by an algorithm to induce misclassification. Several works have proposed methods for defending against adversarial images, however these defenses have shown to be inefficient, since they have facilitated the understanding of their internal operation by attackers. Thus, this paper proposes a defense called MultiMagNet, which randomly incorporates at runtime multiple defense components, in an attempt to introduce an expanded form of non-deterministic behavior so as to hinder evasions by adversarial attacks. Experiments performed on MNIST and CIFAR-10 datasets prove that MultiMagNet can protect classification models from adversarial images generated by the main existing attacks algorithms.

Paper Citation

in Harvard Style

Machado G., Goldschmidt R. and Silva E. (2019). MultiMagNet: A Non-deterministic Approach based on the Formation of Ensembles for Defending Against Adversarial Images.In Proceedings of the 21st International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-372-8, pages 307-318. DOI: 10.5220/0007714203070318

in Bibtex Style

@conference{iceis19,
author={Gabriel Machado and Ronaldo Goldschmidt and Eugênio Silva},
title={MultiMagNet: A Non-deterministic Approach based on the Formation of Ensembles for Defending Against Adversarial Images},
booktitle={Proceedings of the 21st International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2019},
pages={307-318},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007714203070318},
isbn={978-989-758-372-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 21st International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - MultiMagNet: A Non-deterministic Approach based on the Formation of Ensembles for Defending Against Adversarial Images
SN - 978-989-758-372-8
AU - Machado G.
AU - Goldschmidt R.
AU - Silva E.
PY - 2019
SP - 307
EP - 318
DO - 10.5220/0007714203070318