Protecting Classifiers From Attacks

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Gallego, Victor, Naveiro, Roi, Redondo, Alberto, Insua, David Rios, Ruggeri, Fabrizio
Format: Preprint
Publié: 2020
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913932728336384
author Gallego, Victor
Naveiro, Roi
Redondo, Alberto
Insua, David Rios
Ruggeri, Fabrizio
author_facet Gallego, Victor
Naveiro, Roi
Redondo, Alberto
Insua, David Rios
Ruggeri, Fabrizio
contents In multiple domains such as malware detection, automated driving systems, or fraud detection, classification algorithms are susceptible to being attacked by malicious agents willing to perturb the value of instance covariates to pursue certain goals. Such problems pertain to the field of adversarial machine learning and have been mainly dealt with, perhaps implicitly, through game-theoretic ideas with strong underlying common knowledge assumptions. These are not realistic in numerous application domains in relation to security and business competition. We present an alternative Bayesian decision theoretic framework that accounts for the uncertainty about the attacker's behavior using adversarial risk analysis concepts. In doing so, we also present core ideas in adversarial machine learning to a statistical audience. A key ingredient in our framework is the ability to sample from the distribution of originating instances given the, possibly attacked, observed ones. We propose an initial procedure based on approximate Bayesian computation usable during operations; within it, we simulate the attacker's problem taking into account our uncertainty about his elements. Large-scale problems require an alternative scalable approach implementable during the training stage. Globally, we are able to robustify statistical classification algorithms against malicious attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2004_08705
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Protecting Classifiers From Attacks
Gallego, Victor
Naveiro, Roi
Redondo, Alberto
Insua, David Rios
Ruggeri, Fabrizio
Machine Learning
Cryptography and Security
Computation
In multiple domains such as malware detection, automated driving systems, or fraud detection, classification algorithms are susceptible to being attacked by malicious agents willing to perturb the value of instance covariates to pursue certain goals. Such problems pertain to the field of adversarial machine learning and have been mainly dealt with, perhaps implicitly, through game-theoretic ideas with strong underlying common knowledge assumptions. These are not realistic in numerous application domains in relation to security and business competition. We present an alternative Bayesian decision theoretic framework that accounts for the uncertainty about the attacker's behavior using adversarial risk analysis concepts. In doing so, we also present core ideas in adversarial machine learning to a statistical audience. A key ingredient in our framework is the ability to sample from the distribution of originating instances given the, possibly attacked, observed ones. We propose an initial procedure based on approximate Bayesian computation usable during operations; within it, we simulate the attacker's problem taking into account our uncertainty about his elements. Large-scale problems require an alternative scalable approach implementable during the training stage. Globally, we are able to robustify statistical classification algorithms against malicious attacks.
title Protecting Classifiers From Attacks
topic Machine Learning
Cryptography and Security
Computation
url https://arxiv.org/abs/2004.08705