• Home
  • Search
  • Adversarial-Inspired Backdoor Defense via Bridging Backdoor and Adversarial Attacks
  • Cite Icon6
  • https://doi.org/10.1609/aaai.v39i9.33030Copy DOI Icon

Adversarial-Inspired Backdoor Defense via Bridging Backdoor and Adversarial Attacks

  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

Backdoor attacks and adversarial attacks are two major security threats to deep neural networks (DNNs), with the former one is a training-time data poisoning attack that aims to implant backdoor triggers into models by injecting trigger patterns into training samples, and the latter one is a testing-time attack trying to generate adversarial examples (AEs) from benign images to mislead a well-trained model. While previous works generally treat these two attacks separately, the inherent connection between these two attacks is rarely explored. In this paper, we focus on bridging backdoor and adversarial attacks and observe two intriguing phenomena when applying adversarial attacks on an infected model implanted with backdoors: 1) the sample is harder to be turned into an AE when the trigger is presented; 2) the AEs generated from backdoor samples are highly likely to be predicted as its true labels. Inspired by these observations, we proposed a novel backdoor defense method, dubbed Adversarial-Inspired Backdoor Defense (AIBD), to isolate the backdoor samples by leveraging a progressive top-q scheme and break the correlation between backdoor samples and their target labels using adversarial labels. Through extensive experiments on various datasets against six state-of-the-art backdoor attacks, the AIBD-trained models on poisoned data demonstrate superior performance over the existing defense methods.

Similar Papers
  • Research Article
  • Citations9

Towards Unified Robustness Against Both Backdoor and Adversarial Attacks.

  • Dec 01, 2024
  • IEEE transactions on pattern analysis and machine intelligence
  • Zhenxing Niu +4
  • Conference Article
  • Citations4

Defending Deep Learning Based Anomaly Detection Systems Against White-Box Adversarial Examples and Backdoor Attacks

  • Nov 12, 2020
  • Khaled Alrawashdeh +1
  • Conference Article
  • Citations5

Optimizing Deep Learning Based Intrusion Detection Systems Defense Against White-Box and Backdoor Adversarial Attacks Through a Genetic Algorithm

  • Oct 13, 2020
  • Khaled Alrawashdeh +1
  • Conference Article
  • Citations28

Vulnerabilities of Deep Learning-Driven Semantic Communications to Backdoor (Trojan) Attacks

  • Mar 22, 2023
  • Yalin E Sagduyu +3
  • Research Article
  • Citations12

Can We Mitigate Backdoor Attack Using Adversarial Detection Methods?

  • Jul 01, 2023
  • IEEE Transactions on Dependable and Secure Computing
  • Kaidi Jin +6
  • Research Article

Black-Box Adversarial Defense Based on Image Decomposition and Reconstruction

  • Jan 01, 2025
  • IEEE Transactions on Multimedia
  • Jimiao Yu +6
  • PDF
  • Conference Article
  • Citations117

Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer

  • Jan 01, 2021
  • Fanchao Qi +5
  • Conference Article
  • Citations1

A Defense Method against Backdoor Attacks in Neural Networks Using an Image Repair Technique

  • Nov 01, 2022
  • Jiangtao Chen +5
  • Single Book

No Time to Spare: Adversarial Machine Learning at Training and Inference Time

  • Jan 12, 2026
  • Xiaoyun Xu
  • Video Transcripts

Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer

  • Nov 08, 2021
  • Underline Science Inc.
  • Yangyi Chen
  • Conference Article
  • Citations42

Transferable Graph Backdoor Attack

  • Oct 26, 2022
  • Shuiqiao Yang +7
  • Research Article
  • Citations60

One-to-N & N-to-One: Two Advanced Backdoor Attacks Against Deep Learning Models

  • Oct 02, 2020
  • IEEE Transactions on Dependable and Secure Computing
  • Mingfu Xue +3
  • Conference Article
  • Citations10

Test-Time Detection of Backdoor Triggers for Poisoned Deep Neural Networks

  • May 23, 2022
  • Xi Li +3
  • Research Article
  • Citations3

A frequency-injection backdoor attack against DNN-Based finger vein verification

  • Jun 10, 2024
  • Computers & Security
  • Huijie Zhang +2
  • Research Article
  • Citations24

Universal backdoor attack on deep neural networks for malware detection

  • May 26, 2023
  • Applied Soft Computing
  • Yunchun Zhang +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.