检索结果-内蒙古大学图书馆

IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： Yunhui Zhu Jingyi Cao Bo Liu Tingxi Chen Rong Xie Li Song Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Shanghai China School of Computer Science University of Technology Sydney Sydney Australia Cooperative Medianet Innovation Center (CMIC) Shanghai Jiao Tong University Shanghai China

ISBN: (数字)9798350364262

ISBN: (纸本)9798350364279

With the rise of deep learning and the widespread use of face recognition, face image privacy has become a critical research issue. Face de-identification is acknowledged as effective for protecting identity privacy. As media formats diversify, it is imperative to extend privacy protection to videos. Addressing the core problem of identity consistency between frames, we propose a video de-identification approach based on the diffusion model. We disentangle video features with a diffusion autoencoder, where the identity and motion features are encoded into high-level semantic spaces while background and other facial identity-independent features into low-dimensional random subcodes. The unified time-independent identity representation is used to achieve coherent video de-identification results. Compared to existing methods, our proposed approach demonstrates superior performance in terms of privacy protection effectiveness, identity consistency between frames, and utility.

关键词： Privacy Face recognition Semantics Video sequences Media Diffusion models Multimedia communication

来源：评论

学校读者我要写书评

暂无评论

An efficient and secure arbitrary N-party quantum key agreement protocol using Bell states

arXiv

引用

arXiv 2023年

作者： Liu, Wen-Jie Xu, Yong Yang, Ching-Nung Gao, Pei-Pei Yu, Wen-Bin Nanjing210044 China Jiangsu Engineering Center of Network Monitoring Nanjing University of Information Science & Technology Nanjing210044 China School of Computer and Software Nanjing University of Information Science & Technology Nanjing210044 China Department of Computer Science and Information Engineering National Dong Hwa University Hualien974 Taiwan

Two quantum key agreement protocols using Bell states and Bell measurement were recently proposed by Shukla et al. [Quantum Inf. Process. 13, 2391(2014)]. However, Zhu et al. pointed out that there are some security flaws and proposed an improved version[Quantum Inf. Process. 14, 4245(2015)]. In this study, we will show Zhu et al.’s improvement still exists some security problems, and its efficiency is not high enough. For solving these problems, we utilize four Pauli operations {I, Z, X, Y } to encode two bits instead of the original two operations {I, X} to encode one bit, and then propose an efficient and secure arbitrary N-party quantum key agreement protocol. In the protocol, the channel checking with decoy single photons is introduced to avoid the eavesdropper’s flip attack, and a post-measurement mechanism is used to prevent against the collusion attack. The security analysis shows the present protocol can guarantee the correctness, security, privacy and fairness of quantum key agreement. © 2023, CC BY.

关键词： Photons

来源：评论

学校读者我要写书评

暂无评论

A residual attention-based privacy-preserving biometrics model of transcriptome prediction from genome

A residual attention-based privacy-preserving biometrics mod...

引用

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Cheng Tian Song Liu Jinbao Li Guangchen Wang Luyue Kong Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Department of Medicine Shandong Medical College Jinan China

ISBN: (数字)9798350381993

ISBN: (纸本)9798350382006

Transcriptome prediction from genetic variation data is an important task in the privacy-preserving and biometrics field, which can better protect genomic data and achieve biometric recognition through transcriptome. Many transcriptome prediction methods have achieved good accuracy from genetic variation data. However, these traditional transcriptome prediction methods have the problems of linear assumption, overfitting, expose personal privacy, and extensive manual optimization. To solve these shortcomings, we propose an attention-based transcriptome prediction model from genetic variation named RATPM that improves the accuracy of transcriptome prediction and protects participant genomic data. In RATPM, we introduce and improve the deep learning model with multi-head self-attention into the transcriptome prediction stage of Predixcan, which uncovers the non-linear relationship between genetic variation and transcriptome. Moreover, we introduce a residual attention module to generate attention-aware features and extract more accurate features at different levels from genetic variation. Furthermore, we introduce the BERT pre-training module to encode genetic variation fully utilizing their contextual information. Our research enables scientific institutions to publish only predicted transcriptomic data for biometric purposes, thus protecting the genomic information of the subjects. Finally, we evaluated our model on the 1000 Genomes and Geuvadis projects datasets to compare with other baseline models.

关键词： Privacy Biometrics (access control) Biological system modeling Genomics Predictive models Feature extraction Gene expression

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Model Integrating Multi-Omic and Topological Information of PPI network for Drug Synergism Prediction

A Hybrid Model Integrating Multi-Omic and Topological Inform...

引用

IEEE International Conference on Research, Innovation and Vision for the Future

作者： Giang T.T. Nguyen Kien T. Phuong Khanh Nguyen-Trong Duc-Hau Le Network Information Center Hanoi University of Science and Technology Hanoi Vietnam School of Information and Communication Technology Hanoi University of Science and Technology Hanoi Vietnam Faculty of Information Technology Telecommunications Institute of Technology Hanoi Vietnam Faculty of Computer Science and Engineering Thuyloi University Hanoi Vietnam

Combination therapies in precision medicine promise to enhance treatment efficacy and combat drug resistance. Current drug combination prediction methods struggle to extract valuable insights from multi-omics cell line data and integrate them with protein-protein interaction (PPI) network information. To address these challenges, we introduce AE-XGBSynergy, a novel approach. AE-XGBSynergy integrates multi-omics cell line data and drug-cell line features derived from the PPI network's topological characteristics to predict drug synergy combinations. It employs the struc2vec algorithm to extract features for drugs and cell lines based on the PPI network's topological characteristics. Additionally, a pre-trained encoder captures abstract representations of high-dimensional cell line data. These representations of drug pairs and cell lines serve as input for an extreme gradient-boosting algorithm to predict drug combinations for cancer cell lines. We assessed our model's performance using two datasets: the O'Neil and DrugCombDB datasets, which provide extensive drug combination information. Our experimental results show that AE-XGBSynergy excels at integrating multi-omics data to predict drug synergies, with genomic data playing a particularly significant role. Notably, when compared to NEXGB, a hybrid model that includes drug and cell line features using a PPI network, our approach outperforms in terms of accuracy (ACC), AVC-ROC, AUC-PR, Precision, and F1-score.

关键词：

来源：评论

学校读者我要写书评

暂无评论

GNMS: A novel method for model stealing based on GAN

GNMS: A novel method for model stealing based on GAN

引用

International Conference on Advanced Cloud and Big Data (CBD)

作者： Moxuan Zeng Yangzhong Wang Yangming Zhang Jun Niu Yuqing Zhang College of Cyberspace Security Hainan University Haikou National Computer Network Intrusion Protection Center University of Chinese Academy of Sciences Beijing School of Computer Science and Technology Xidian University Xian School of Cyber Engineering Xidian University Xian

Many well-performing models are currently deployed on the cloud to provide machine Learning as a service (MLaaS). However, these models are susceptible to Model Stealing Attacks, where attackers can access the model’s functionality, parameters, and internal structure in a black-box. As a result, data-free model stealing methods have gained popularity due to their higher accuracy and not requiring real data. Previous data-free model stealing methods have mainly focused on single scenarios and limited model and dataset variations. In this paper, we introduce a novel generalized network model Stealing method (GNMS), which is suitable for both benchmark and transfer models, achieving high model stealing accuracy across various scenarios. We pre-train generative adversarial network (GAN) using publicly available datasets and efficiently steal model functionality by training a student model with the pre-trained generator and the discriminator. Adversarial samples and the generated image dataset are also used to explore the model’s decision boundaries. During the training of the clone model, we train two clone models to minimize the differences with the target model further. We employ a contrastive learning approach to encourage the models to learn meaningful feature representations by distinguishing between similar and dissimilar data points, thereby enhancing the model’s accuracy. We achieve a model stealing accuracy of 73.02% and 72.93% on more complex datasets CIFAR100 and Caltech101. Surpass the latest DisGUIDE by 3.55% and 2.61%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Logits Poisoning Attack in Federated Distillation

arXiv

引用

arXiv 2024年

作者： Tang, Yuhan Wu, Zhiyuan Gao, Bo Wen, Tian Wang, Yuwei Sun, Sheng Engineering Research Center of Network Management Technology for High-Speed Railway Ministry of Education School of Computer and Information Technology Bejing Jiaotong University Beijing China Collaborative Innovation Center of Railway Traffic Safety Beijing Jiaotong University Beijing China Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

Federated Distillation (FD) is a novel and promising distributed machine learning paradigm, where knowledge distillation is leveraged to facilitate a more efficient and flexible cross-device knowledge transfer in federated learning. By optimizing local models with knowledge distillation, FD circumvents the necessity of uploading large-scale model parameters to the central server, simultaneously preserving the raw data on local clients. Despite the growing popularity of FD, there is a noticeable gap in previous works concerning the exploration of poisoning attacks within this framework. This can lead to a scant understanding of the vulnerabilities to potential adversarial actions. To this end, we introduce FDLA, a poisoning attack method tailored for FD. FDLA manipulates logit communications in FD, aiming to significantly degrade model performance on clients through misleading the discrimination of private samples. Through extensive simulation experiments across a variety of datasets, attack scenarios, and FD configurations, we demonstrate that LPA effectively compromises client model accuracy, outperforming established baseline algorithms in this regard. Our findings underscore the critical need for robust defense mechanisms in FD settings to mitigate such adversarial threats. Copyright © 2024, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Stacked Sparse Autoencoder for Audio Object Coding 1

引用

27th International Conference on MultiMedia Modeling, MMM 2021

作者： Wu, Yulin Hu, Ruimin Wang, Xiaochen Hu, Chenhao Li, Gang National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University Wuhan China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China Collaborative Innovation Center of Geospatial Technology Wuhan China

ISBN: (数字)9783030678326

ISBN: (纸本)9783030678319

Compared with channel-based audio coding, the object-based audio coding has a definite advantage in meeting the user’s demands of personalized control. However, in the conventional Spatial Audio Object Coding (SAOC), each frame is divided into 28 sub-bands. All frequency points in one sub-band share the common parameter. Under the SAOC framework, the bitrate can be saved, but aliasing distortion is prone to occur, which will influence the listening experience of audiences. In order to obtain higher perceptual quality, we propose a Stacked Sparse Autoencoder (SSAE) pipeline as overlapped modules. Each module extracted the efficient feature of side information from its preceding module. Then we can reduce the dimensionality of side information parameters for saving bitrate, and well reconstruct audio objects, thereby providing favorable auditory perception. Compared with conventional SAOC, TS-SAOC, and SVD-SAOC, both objective and subjective results show that the proposed method can achieve the best sound quality of the output signal at the same bitrate. © 2021, Springer Nature Switzerland AG.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Semisupervised Encrypted Traffic Identification Based on Auxiliary Classification Generative Adversarial network

引用

computer Systems Science & engineering 2021年第12期39卷 373-390页

作者： Jiaming Mao Mingming Zhang Mu Chen Lu Chen Fei Xia Lei Fan ZiXuan Wang Wenbing Zhao State Grid Jiangsu Electric Power Co. Ltd.Information and Telecommunication BranchNanJing210003China Institute of Information and Communication Global Energy Interconnection Research InstituteState Grid Key Laboratory of Information and Network SecurityNanjing210003China Engineering Research Center of Post Big Data Technology and Application of Jiangsu Province Research and Development Center of Post Industry Technology of the State Posts Bureau(Internet of Things Technology)Broadband Wireless Communication Technology Engineering Research Center of the Ministry of EducationNanjing University of Posts and TelecommunicationsNanjing210003China Department of Electrical Engineering and Computer Science Cleveland State UniversityCleveland44115USA

The rapidly increasing popularity of mobile devices has changed the methods with which people access various network services and increased net-work traffic *** the past few decades,network traffic identification has been a research hotspot in the field of network management and security ***,as more network services use encryption technology,network traffic identification faces many *** classic machine learning methods can solve many problems that cannot be solved by port-and payload-based methods,manually extract features that are frequently updated is time-consuming and *** learning has good automatic feature learning capabilities and is an ideal method for network traffic identification,particularly encrypted traffic identification;Existing recognition methods based on deep learning primarily use supervised learning methods and rely on many labeled ***,in real scenarios,labeled samples are often difficult to *** paper adjusts the structure of the auxiliary classification generation adversarial network(ACGAN)so that it can use unlabeled samples for training,and use the wasserstein distance instead of the original cross entropy as the loss function to achieve semisupervised *** results show that the identification accuracy of ISCX and USTC data sets using the proposed method yields markedly better performance when the number of labeled samples is small compared to that of convolutional neural network(CNN)based classifier.

关键词： Encrypted traffic recognition deep learning generative adversarial network traffic classification semisupervised learning

来源：评论

学校读者我要写书评

暂无评论

Robust Imagined Speech Production from Electrocorticography with Adaptive Frequency Enhancement

Robust Imagined Speech Production from Electrocorticography ...

引用

International Joint Conference on Neural networks (IJCNN)

作者： Xu Xu Chong Fu Junxin Chen Gwanggil Jeon David Camacho School of Computer Science and Engineering Northeastern University Shenyang China School of Software Dalian University of Technology Dalian China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Northeastern University Shenyang China Engineering Research Center of Security Technology of Complex Network System Ministry of Education Shenyang China College of Information Technology Incheon National University Incheon Korea Department of Computer Systems Engineering Universidad Politécnica de Madrid Madrid Spain

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Imagined speech production with electrocorticography (ECoG) plays a crucial role in brain-computer interface system. A challenging issue is the great variation underlying the frequency bands of the ECoG signals’ encode information, which makes current methods difficult to generate imagined speech with stable quality among different persons. To this end, we propose a robust model to generate high-quality imagined speech from ECoG. A frequency enhancement branch is first designed to adaptively modulate the frequency information, whose product is fed into the following multi-scale channel attention module for robust feature extraction and fusion. By incorporating both the mel-spectrum and audio as training constraints, a multi-constraint decoder branch is finally constructed for imagined speech production. The performance of our model is evaluated on a high-quality dateset, i.e, Single Word Production Dutch-iBIDS. It yields Pearson correlation scores that are all above 0.8, and the standard deviationsare are all below 0.2 in different volunteers. Experimental results demonstrate that our model is effective and robust for ECoG based imagined speech production, and has advantages over peer methods.

关键词： Training Adaptation models Frequency modulation Correlation Neuroscience Production Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Effects of Optimized Operating Parameters on Combustion Characteristics and NO_(x)Emissions of a Burner based on Orthogonal Analysis

引用

Journal of Thermal Science 2021年第4期30卷 1212-1223页

作者： XU Qian AKKURT Nevzat YANG Gang ZHU Lidong SHI Kejian WANG Kang ZOU Zhenwei LIU Zhihui WANG Jiulong DU Zhiwei School of Energy and Environmental Engineering University of Science and Technology BeijingBeijing 100083China Beijing Key Laboratory for Energy Conservation and Emission Reduction for Metallurgical Industry Beijing 100083China Department of Mechanical Engineering Munzur UniversityTunceli 62000Turkey Department of Big Data Technology and Application Development Computer Network Information CenterChinese Academy of SciencesBeijing 100190China AVIC Xinxiang Aviation Industry(group)CO. LTDHenan 453002China

To optimize the structure of the burner,improve the combustion performance,and reduce the emission of NO_(x),a self-circulating low NO_(x)combustion technology was used to design a new type of flue gas self-circulating low NO_(x)*** on previous research on the numerical model of combustion and the composition of mixed gas on combustion and NO_(x)emissions,the effect of various factors on the ejection coefficient of the flue gas self-circulating structure was analyzed using the orthogonal test method,and the burner operating parameters,such as preheating temperature and excess air coefficient,were deeply studied through the three-dimensional finite element numerical model in this *** results show that the diameter ratio of the nozzle and the length of the cylindrical section of the flue gas self-circulating structure have great influence on its ejection and mixing *** optimal ejection coefficient was ***,the amount of NO_(x)emissions greatly increased from 6.23×10^(-6)(volume fraction)at the preheating temperature 973 K to 3.5×10^(-3)at preheating temperature 1573 *** the excess air coefficient decreased from 1.2 to 1,the maximum combustion temperature decreased from 2036.3 K to 1954.22 K,and the NO_(x)emissions decreased from 352.29×10^(-6)to 159.73×10^(-6).

关键词： preheating temperature excess air coefficient self-circulating flue gas low oxygen combustion low-NO_(x)emissions

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：