paper-with-me

Papers

Explore the Effect of Data Selection on Poison Efficiency in Backdoor Attacks

2023-10-15 · Ziqiang Li, Pengfei Xia, Hong Sun, Yueqi Zeng, Wei zhang, Bin Li

As the number of parameters in Deep Neural Networks (DNNs) scales, the thirst for training data also increases. To save costs, it has become common for users and enterprises to delegate time-consuming data collection to third parties. Unfortunately, recent research has shown that this practice raises the risk of DNNs being exposed to backdoor attacks. Specifically, an attacker can maliciously control the behavior of a trained model by poisoning a small portion of the training data. In this study, we focus on improving the poisoning efficiency of backdoor attacks from the sample selection perspective. The existing attack methods construct such poisoned samples by randomly selecting some clean data from the benign set and then embedding a trigger into them. However, this random selection strategy ignores that each sample may contribute differently to the backdoor injection, thereby reducing the poisoning efficiency. To address the above problem, a new selection strategy named Improved Filtering and Updating Strategy (FUS++) is proposed. Specifically, we adopt the forgetting events of the samples to indicate the contribution of different poisoned samples and use the curvature of the loss surface to analyses the effectiveness of this phenomenon. Accordingly, we combine forgetting events and curvature of different samples to conduct a simple yet efficient sample selection strategy. The experimental results on image classification (CIFAR-10, CIFAR-100, ImageNet-10), text classification (AG News), audio classification (ESC-50), and age regression (Facial Age) consistently demonstrate the effectiveness of the proposed strategy: the attack performance using FUS++ is significantly higher than that using random selection for the same poisoning ratio.

📄 PDF Abstract BibTeX arXiv:2310.09744

Code (0)

등록된 구현이 없습니다.

Tasks

Audio Classificationimage-classificationImage Classificationtext-classificationText Classification

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

A Proxy Attack-Free Strategy for Practically Improving the Poisoning Efficiency in Backdoor Attacks

2023-06-14 · Ziqiang Li, Hong Sun, Pengfei Xia, Beihao Xia 외

Poisoning efficiency is crucial in poisoning-based backdoor attacks, as attackers aim to minimize the number of poisoning samples while maximizing attack efficacy. Recent studies have sought to enhance poisoning efficien…

Active LearningBackdoor Attack

Data-Efficient Backdoor Attacks

2022-04-22 · Pengfei Xia, Ziqiang Li, Wei zhang, Bin Li

Recent studies have proven that deep neural networks are vulnerable to backdoor attacks. Specifically, by mixing a small number of poisoned samples into the training set, the behavior of the trained model can be maliciou…

EAB-FL: Exacerbating Algorithmic Bias through Model Poisoning Attacks in Federated Learning

2024-10-02 · Syed Irfan Ali Meerza, Jian Liu

Federated Learning (FL) is a technique that allows multiple parties to train a shared model collaboratively without disclosing their private data. It has become increasingly popular due to its distinct privacy advantages…

FairnessFederated LearningModel Poisoning

Boosting Backdoor Attack with A Learnable Poisoning Sample Selection Strategy

2023-07-14 · Zihao Zhu, Mingda Zhang, Shaokui Wei, Li Shen 외

Data-poisoning based backdoor attacks aim to insert backdoor into models by manipulating training datasets without controlling the training process of the target model. Existing attack methods mainly focus on designing t…

Backdoor AttackData Poisoning

A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers

2025-09-24 · Zhixiao Wu, Yao Lu, Jie Wen, Hao Sun 외 arxiv

Poison-only Clean-label Backdoor Attacks aim to covertly inject attacker-desired behavior into DNNs by merely poisoning the dataset without changing the labels. To effectively implant a backdoor, multiple \textbf{trigger…