paper-with-me

홈 › Papers

A Synergetic Attack against Neural Network Classifiers combining Backdoor and Adversarial Examples

2021-09-03 · Guanxiong Liu, Issa Khalil, Abdallah Khreishah, NhatHai Phan

In this work, we show how to jointly exploit adversarial perturbation and model poisoning vulnerabilities to practically launch a new stealthy attack, dubbed AdvTrojan. AdvTrojan is stealthy because it can be activated only when: 1) a carefully crafted adversarial perturbation is injected into the input examples during inference, and 2) a Trojan backdoor is implanted during the training process of the model. We leverage adversarial noise in the input space to move Trojan-infected examples across the model decision boundary, making it difficult to detect. The stealthiness behavior of AdvTrojan fools the users into accidentally trust the infected model as a robust classifier against adversarial examples. AdvTrojan can be implemented by only poisoning the training data similar to conventional Trojan backdoor attacks. Our thorough analysis and extensive experiments on several benchmark datasets show that AdvTrojan can bypass existing defenses with a success rate close to 100% in most of our experimental scenarios and can be extended to attack federated learning tasks as well.

📄 PDF Abstract BibTeX arXiv:2109.01275

Code (0)

등록된 구현이 없습니다.

Tasks

Federated LearningModel Poisoning

Similar Papers 제목 키워드 기반

Detecting Backdoor Attacks Against Point Cloud Classifiers

2021-10-20 · Zhen Xiang, David J. Miller, Siheng Chen, Xi Li 외

Backdoor attacks (BA) are an emerging threat to deep neural network classifiers. A classifier being attacked will predict to the attacker's target class when a test sample from a source class is embedded with the backdoo…

Autonomous Driving

Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifiers

2022-02-11 · Limin Yang, Zhi Chen, Jacopo Cortellazzi, Feargus Pendlebury 외

Malware classifiers are subject to training-time exploitation due to the need to regularly retrain using samples collected from the wild. Recent work has demonstrated the feasibility of backdoor attacks against malware c…

Backdoor Attack

Explanation-Guided Backdoor Poisoning Attacks Against Malware Classifiers

2020-03-02 · Giorgio Severi, Jim Meyer, Scott Coull, Alina Oprea

Training pipelines for machine learning (ML) based malware classification often rely on crowdsourced threat feeds, exposing a natural attack injection point. In this paper, we study the susceptibility of feature-based ML…

BIG-bench Machine LearningGeneral ClassificationMalware Classification

Detection of Backdoors in Trained Classifiers Without Access to the Training Set

2019-08-27 · Zhen Xiang, David J. Miller, George Kesidis

Recently, a special type of data poisoning (DP) attack targeting Deep Neural Network (DNN) classifiers, known as a backdoor, was proposed. These attacks do not seek to degrade classification accuracy, but rather to have …

Anomaly DetectionData PoisoningUnsupervised Anomaly Detection

Detecting Backdoors in Deep Text Classifiers

2022-10-11 · You Guo, Jun Wang, Trevor Cohn

Deep neural networks are vulnerable to adversarial attacks, such as backdoor attacks in which a malicious adversary compromises a model during training such that specific behaviour can be triggered at test time by attach…

Data Poisoningtext-classificationText Classification