paper-with-me

홈 › Papers

BadActs: A Universal Backdoor Defense in the Activation Space

2024-05-18 · Biao Yi, Sishuo Chen, Yiming Li, Tong Li, Baolei Zhang, Zheli Liu

Backdoor attacks pose an increasingly severe security threat to Deep Neural Networks (DNNs) during their development stage. In response, backdoor sample purification has emerged as a promising defense mechanism, aiming to eliminate backdoor triggers while preserving the integrity of the clean content in the samples. However, existing approaches have been predominantly focused on the word space, which are ineffective against feature-space triggers and significantly impair performance on clean data. To address this, we introduce a universal backdoor defense that purifies backdoor samples in the activation space by drawing abnormal activations towards optimized minimum clean activation distribution intervals. The advantages of our approach are twofold: (1) By operating in the activation space, our method captures from surface-level information like words to higher-level semantic concepts such as syntax, thus counteracting diverse triggers; (2) the fine-grained continuous nature of the activation space allows for more precise preservation of clean content while removing triggers. Furthermore, we propose a detection module based on statistical information of abnormal activations, to achieve a better trade-off between clean accuracy and defending performance.

📄 PDF Abstract BibTeX arXiv:2405.11227

Code (1)

clearloveclearlove/BadActs 공식 구현 pytorch

Tasks

backdoor defense

Similar Papers 제목 키워드 기반

Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attack

2024-05-25 · Mingli Zhu, Siyuan Liang, Baoyuan Wu

Deep neural networks face persistent challenges in defending against backdoor attacks, leading to an ongoing battle between attacks and defenses. While existing backdoor defense strategies have shown promising performanc…

Adversarial Attackbackdoor defenseContrastive Learningimage-classification+1

FLARE: Toward Universal Dataset Purification against Backdoor Attacks

2024-11-29 · Linshan Hou, Wei Luo, Zhongyun Hua, Songhua Chen 외

Deep neural networks (DNNs) are susceptible to backdoor attacks, where adversaries poison datasets with adversary-specified triggers to implant hidden backdoors, enabling malicious manipulation of model predictions. Data…

All

Securing Federated Learning against Backdoor Threats with Foundation Model Integration

2024-10-23 · Xiaohuan Bi, Xi Li

Federated Learning (FL) enables decentralized model training while preserving privacy. Recently, the integration of Foundation Models (FMs) into FL has enhanced performance but introduced a novel backdoor attack mechanis…

Backdoor AttackFederated Learning

Defending Deep Regression Models against Backdoor Attacks

2024-11-07 · Lingyu Du, Yupei Liu, Jinyuan Jia, Guohao Lan

Deep regression models are used in a wide variety of safety-critical applications, but are vulnerable to backdoor attacks. Although many defenses have been proposed for classification models, they are ineffective as they…

regression

The "Beatrix'' Resurrections: Robust Backdoor Detection via Gram Matrices

2022-09-23 · Wanlun Ma, Derui Wang, Ruoxi Sun, Minhui Xue 외

Deep Neural Networks (DNNs) are susceptible to backdoor attacks during training. The model corrupted in this way functions normally, but when triggered by certain patterns in the input, produces a predefined target label…

valid