Adversarial Examples - A Complete Characterisation of the Phenomenon
Computer Vision and Pattern Recognition
2019-02-19 v2 Cryptography and Security
Machine Learning
Neural and Evolutionary Computing
Abstract
We provide a complete characterisation of the phenomenon of adversarial examples - inputs intentionally crafted to fool machine learning models. We aim to cover all the important concerns in this field of study: (1) the conjectures on the existence of adversarial examples, (2) the security, safety and robustness implications, (3) the methods used to generate and (4) protect against adversarial examples and (5) the ability of adversarial examples to transfer between different machine learning models. We provide ample background information in an effort to make this document self-contained. Therefore, this document can be used as survey, tutorial or as a catalog of attacks and defences using adversarial examples.
Cite
@article{arxiv.1810.01185,
title = {Adversarial Examples - A Complete Characterisation of the Phenomenon},
author = {Alexandru Constantin Serban and Erik Poll and Joost Visser},
journal= {arXiv preprint arXiv:1810.01185},
year = {2019}
}