De-Pois: An Attack-Agnostic Defense against Data Poisoning Attacks

Machine learning techniques have been widely applied to various applications. However, they are potentially vulnerable to data poisoning attacks, where sophisticated attackers can disrupt the learning procedure by injecting a fraction of malicious samples into the training dataset. Existing defense...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on information forensics and security 2021, Vol.16, p.3412-3425
Main Authors: Chen, Jian, Zhang, Xuxin, Zhang, Rui, Wang, Chen, Liu, Ling
Format: Article
Language:English
Subjects:
Algorithms
attack-agnostic defense
Computational modeling
Data models
data poisoning attack
Datasets
Defense
Explicit knowledge
generative adversarial network
Generative adversarial networks
Machine learning
Predictive models
Task analysis
Testing
Training
Training data
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Machine learning techniques have been widely applied to various applications. However, they are potentially vulnerable to data poisoning attacks, where sophisticated attackers can disrupt the learning procedure by injecting a fraction of malicious samples into the training dataset. Existing defense techniques against poisoning attacks are largely attack-specific : they are designed for one specific type of attacks but do not work for other types, mainly due to the distinct principles they follow. Yet few general defense strategies have been developed. In this paper, we propose De-Pois, an attack-agnostic defense against poisoning attacks. The key idea of De-Pois is to train a mimic model the purpose of which is to imitate the behavior of the target model trained by clean samples. We take advantage of Generative Adversarial Networks (GANs) to facilitate informative training data augmentation as well as the mimic model construction. By comparing the prediction differences between the mimic model and the target model, De-Pois is thus able to distinguish the poisoned samples from clean ones, without explicit knowledge of any ML algorithms or types of poisoning attacks. We implement four types of poisoning attacks and evaluate De-Pois with five typical defense methods on different realistic datasets. The results demonstrate that De-Pois is effective and efficient for detecting poisoned data against all the four types of poisoning attacks, with both the accuracy and F1-score over 0.9 on average.
ISSN:1556-6013
1556-6021
DOI:10.1109/TIFS.2021.3080522