Back to Search Start Over

A Textual Backdoor Defense Method Based on Deep Feature Classification.

Authors :
Shao, Kun
Yang, Junan
Hu, Pengjiang
Li, Xiaoshuai
Source :
Entropy; Feb2023, Vol. 25 Issue 2, p220, 13p
Publication Year :
2023

Abstract

Natural language processing (NLP) models based on deep neural networks (DNNs) are vulnerable to backdoor attacks. Existing backdoor defense methods have limited effectiveness and coverage scenarios. We propose a textual backdoor defense method based on deep feature classification. The method includes deep feature extraction and classifier construction. The method exploits the distinguishability of deep features of poisoned data and benign data. Backdoor defense is implemented in both offline and online scenarios. We conducted defense experiments on two datasets and two models for a variety of backdoor attacks. The experimental results demonstrate the effectiveness of this defense approach and outperform the baseline defense method. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
10994300
Volume :
25
Issue :
2
Database :
Complementary Index
Journal :
Entropy
Publication Type :
Academic Journal
Accession number :
162117883
Full Text :
https://doi.org/10.3390/e25020220