检索结果-南通市图书馆

Adaptive backdoor attack against Deep Neural Networks

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2023年第9期136卷 2617-2633页

作者： Honglu He Zhiying Zhu Xinpeng Zhang School of Computer Science Fudan UniversityShanghai200433China

In recent years,the number of parameters of deep neural networks(DNNs)has been increasing *** training of DNNs is typically *** a result,many users leverage cloud computing and outsource their training *** computation results in a potential risk called backdoor attack,in which a welltrained DNN would performabnormally on inputs with a certain *** attacks can also be classified as attacks that exploit fake ***,most backdoor attacks design a uniformtrigger for all images,which can be easilydetectedand *** this paper,we propose a novel adaptivebackdoor *** overcome this defect and design a generator to assign a unique trigger for each image depending on its *** achieve this goal,we use a texture complexitymetric to create a specialmask for eachimage,which forces the trigger tobe embedded into the rich texture *** trigger is distributed in texture regions,which makes it invisible to *** the stealthiness of triggers,we limit the range of modification of backdoor models to evade *** show that our method is efficient in multiple datasets,and traditional detectors cannot reveal the existence of a backdoor.

关键词： backdoor attack AI security DNN

An Improved Optimized Model for Invisible backdoor attack Creation Using Steganography

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第7期72卷 1173-1193页

作者： Daniyal M.Alghazzawi Osama Bassam J.Rabie Surbhi Bhatia Syed Hamid Hasan Faculty of Computing and Information Technology King Abdulaziz UniversityJeddah21589Saudi Arabia Department of Information Systems College of Computer Sciences and Information TechnologyKing Faisal UniversitySaudi Arabia

The Deep Neural Networks(DNN)training process is widely affected by backdoor *** backdoor attack is excellent at concealing its identity in the DNN by performing well on regular samples and displaying malicious behavior with data poisoning *** state-of-art backdoor attacks mainly follow a certain assumption that the trigger is sample-agnostic and different poisoned samples use the same *** overcome this problem,in this work we are creating a backdoor attack to check their strength to withstand complex defense strategies,and in order to achieve this objective,we are developing an improved Convolutional Neural Network(ICNN)model optimized using a Gradient-based Optimization(GBO)(ICNN-GBO)*** the ICNN-GBO model,we are injecting the triggers via a steganography and regularization *** are generating triggers using a single-pixel,irregular shape,and different *** performance of the proposed methodology is evaluated using different performance metrics such as attack success rate,stealthiness,pollution index,anomaly index,entropy index,and *** the CNN-GBO model is trained with the poisoned dataset,it will map the malicious code to the target *** proposed scheme’s effectiveness is verified by the experiments conducted on both the benchmark datasets namely CIDAR-10 andMSCELEB 1M *** results demonstrate that the proposed methodology offers significant defense against the conventional backdoor attack detection frameworks such as STRIP and Neutral cleanse.

关键词： Convolutional neural network gradient-based optimization steganography backdoor attack and regularization attack

backdoor attacks on Image Classification Models in Deep Neural Networks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2022年第2期31卷 199-212页

作者： ZHANG Quanxin MA Wencong WANG Yajie ZHANG Yaoyuan SHI Zhiwei LI Yuanzhang Beijing Institute of Technology China Information Technology Security Evaluation Center

Deep neural network(DNN) is applied widely in many applications and achieves state-of-the-art performance. However, DNN lacks transparency and interpretability for users in structure. attackers can use this feature to embed trojan horses in the DNN structure,such as inserting a backdoor into the DNN, so that DNN can learn both the normal main task and additional malicious tasks at the same time. Besides, DNN relies on data set for training. attackers can tamper with training data to interfere with DNN training process, such as attaching a trigger on input data. Because of defects in DNN structure and data, the backdoor attack can be a serious threat to the security of DNN. The DNN attacked by backdoor performs well on benign inputs while it outputs an attacker-specified label on trigger attached inputs. backdoor attack can be conducted in almost every stage of the machine learning pipeline. Although there are a few researches in the backdoor attack on image classification, a systematic review is still rare in this field. This paper is a comprehensive review of backdoor attacks. According to whether attackers have access to the training data, we divide various backdoor attacks into two types: poisoning-based attacks and non-poisoning-based attacks. We go through the details of each work in the timeline, discussing its contribution and deficiencies. We propose a detailed mathematical backdoor model to summary all kinds of backdoor attacks. In the end, we provide some insights about future studies.

关键词： nonpoisoning-based attacks image classification mathematical backdoor model machine learning pipeline backdoor attack deep learning (artificial intelligence) DNN training process image classification models trojan horses deep neural network invasive software computer network security

Active poisoning: efficient backdoor attacks on transfer learning-based brain-computer interfaces

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第8期66卷 211-232页

作者： Xue JIANG Lubin MENG Siyang LI Dongrui WU School of Artificial Intel ligence and Automation Huazhong University of Science and Technology Zhejiang Lab

Transfer learning(TL) has been widely used in electroencephalogram(EEG)-based braincomputer interfaces(BCIs) for reducing calibration efforts. However, backdoor attacks could be introduced through TL. In such attacks, an attacker embeds a backdoor with a specific pattern into the machine learning model. As a result, the model will misclassify a test sample with the backdoor trigger into a prespecified class while still maintaining good performance on benign samples. Accordingly, this study explores backdoor attacks in the TL of EEG-based BCIs, where source-domain data are poisoned by a backdoor trigger and then used in TL. We propose several active poisoning approaches to select source-domain samples, which are most effective in embedding the backdoor pattern, to improve the attack success rate and efficiency. Experiments on four EEG datasets and three deep learning models demonstrate the effectiveness of the approaches. To our knowledge, this is the first study about backdoor attacks on TL models in EEG-based BCIs. It exposes a serious security risk in BCIs, which should be immediately addressed.

关键词： brain-computer interface electroencephalogram transfer learning poisoning attack backdoor attack

DLP:towards active defense against backdoor attacks with decoupled learning process

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Cybersecurity 2024年第1期7卷 122-134页

作者： Zonghao Ying Bin Wu State Key Laboratory of Information Security Institute of Information EngineeringChinese Academy of SciencesBeijingChina School of Cyber Security University of Chinese Academy of SciencesBeijingChina

Deep learning models are well known to be susceptible to backdoor attack,where the attacker only needs to provide a tampered dataset on which the triggers are *** trained on the dataset will passively implant the backdoor,and triggers on the input can mislead the models during *** study shows that the model shows different learning behaviors in clean and poisoned subsets during *** on this observation,we propose a general training pipeline to defend against backdoor attacks *** models can be trained from the unreli-able dataset by decoupling the learning process into three stages,i.e.,supervised learning,active unlearning,and active semi-supervised *** effectiveness of our approach has been shown in numerous experiments across various backdoor attacks and datasets.

关键词： Deep learning backdoor attack Active defense

A Gaussian Noise-Based Algorithm for Enhancing backdoor attacks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第7期80卷 361-387页

作者： Hong Huang Yunfei Wang Guotao Yuan Xin Li School of Computer Science and Engineering Sichuan University of Science&EngineeringYibin644000China

Deep Neural Networks(DNNs)are integral to various aspects of modern life,enhancing work ***-less,their susceptibility to diverse attack methods,including backdoor attacks,raises security *** aim to investigate backdoor attack methods for image categorization tasks,to promote the development of DNN towards higher *** on backdoor attacks currently faces significant challenges due to the distinct and abnormal data patterns of malicious samples,and the meticulous data screening by developers,hindering practical attack *** overcome these challenges,this study proposes a Gaussian Noise-Targeted Universal Adversarial Perturbation(GN-TUAP)*** approach restricts the direction of perturbations and normalizes abnormal pixel values,ensuring that perturbations progress as much as possible in a direction perpendicular to the decision hyperplane in linear *** limits anomalies within the perturbations improves their visual stealthiness,and makes them more challenging for defense methods to *** verify the effectiveness,stealthiness,and robustness of GN-TUAP,we proposed a comprehensive threat *** on this model,extensive experiments were conducted using the CIFAR-10,CIFAR-100,GTSRB,and MNIST datasets,comparing our method with existing state-of-the-art attack *** also tested our perturbation triggers using various defense methods and further experimented on the robustness of the triggers against noise filtering *** experimental outcomes demonstrate that backdoor attacks leveraging perturbations generated via our algorithm exhibit cross-model attack effectiveness and superior ***,they possess robust anti-detection capabilities and maintain commendable performance when subjected to noise-filtering methods.

关键词： Image classification model backdoor attack gaussian distribution Artificial Intelligence(AI)security