Authors

* External authors

Venue

Date

Share

Anti-Backdoor Learning: Training Clean Models on Poisoned Data

Yige Li*

Xixiang Lyu*

Nodens Koren*

Lingjuan Lyu

Bo Li*

Xingjun Ma*

* External authors

NeurIPS-2021

2021

Abstract

Backdoor attack has emerged as a major security threat to deep neural networks(DNNs). While existing defense methods have demonstrated promising results on detecting and erasing backdoor triggers, it is still not clear if measures can be taken to avoid the triggers from being learned into the model in the first place. In this paper, we introduce the concept of anti-backdoor learning, of which the aim is to train clean models on backdoor-poisoned data. We frame the overall learning process as a dual-task of learning the clean portion of data and learning the backdoor portion of data. From this view, we identify two inherent characteristics of backdoor attacks as their weaknesses: 1) the models learn backdoored data at a much faster rate than learning clean data, and the stronger the attack the faster the model converges on backdoored data; and 2) the backdoor task is tied to a specific class (the backdoor target class). Based on these two weaknesses, we propose a general learning scheme, Anti-Backdoor Learning (ABL), to automatically prevent backdoor attacks during training. ABL introduces a two-stage gradient ascent mechanism into standard training to 1) help isolate backdoor examples at an early training stage, and 2) break the correlation between backdoor examples and the target class at a later training stage. Through extensive experiments on multiple benchmark datasets against 10 state-of-the-art attacks, we empirically show that ABL-trained models on backdoor-poisoned data achieve the same performance as they were trained on purely clean data. Code is available athttps://github.com/bboylyg/ABL.

Related Publications

Heterogeneous Graph Node Classification with Multi-Hops Relation Features

ICASSP, 2022
Xiaolong Xu*, Lingjuan Lyu, Hong Jin*, Weiqiang Wang*, Shuo Jia*

In recent years, knowledge graph~(KG) has obtained many achievements in both research and industrial fields. However, most KG algorithms consider node embedding with only structure and node features, but not relation features. In this paper, we propose a novel Heterogeneous …

How to Inject Backdoors with Better Consistency: Logit Anchoring on Clean Data

ICLR, 2022
Zhiyuan Zhang*, Lingjuan Lyu, Weiqiang Wang*, Lichao Sun*, Xu Sun*

Since training a large-scale backdoored model from scratch requires a large training dataset, several recent attacks have considered to inject backdoors into a trained clean model without altering model behaviors on the clean data. Previous work finds that backdoors can be i…

Decision Boundary-aware Data Augmentation for Adversarial Training

TDSC, 2022
Chen Chen, Jingfeng Zhang*, Xilie Xu*, Lingjuan Lyu, Chaochao Chen*, Tianlei Hu*, Gang Chen*

Adversarial training (AT) is a typical method to learn adversarially robust deep neural networks via training on the adversarial variants generated by their natural examples. However, as training progresses, the training data becomes less attackable, which may undermine the …

  • HOME
  • Publications
  • Anti-Backdoor Learning: Training Clean Models on Poisoned Data

JOIN US

Shape the Future of AI with Sony AI

We want to hear from those of you who have a strong desire
to shape the future of AI.