检索结果-内蒙古大学图书馆

SSRN 2022年

作者： Zhang, Yu Liu, Haoliang Xiong, Wei Mi, Siya Bian, Yali School of Computer Science and Engineering The Key Lab of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China School of Cyber Science and Engineering Southeast University Nanjing211189 China Purple Mountain Laboratories Nanjing China Intel Labs United States

Image-text retrieval attempts to find relevant images from a huge corpus of images given a textual description, and vice versa. Existing works usually denote the negative/positive image-text pair with a binary label. They assume that a text is only relevant to the paired images and irrelevant to other images in the corpus. However, using the binary labels cannot reveal the precise similarity between an image and a text. In this paper, we learn the model using soft labels, where each image-text pair is associated with a specific soft label. We generate the Intra-modal Soft Label (ISL) between the cross-modal (image, text) pair by the similarity between the intra-modal (image caption, text) pair. Systematic experiments on public benchmarks, including MSCOCO and Flickr30K, demonstrate the effectiveness of the proposed method, which achieves a recall@1 improvement of approximately 1%-3% compared to the baseline method. © 2022, The Authors. All rights reserved.

关键词： information retrieval

来源：评论

学校读者我要写书评

暂无评论

NCAGC: A Neighborhood Contrast Framework for Attributed Graph Clustering

arXiv

引用

arXiv 2022年

作者： Wang, Tong Yang, Guanyu He, Qijia Zhang, Zhenquan Wu, Junhua The Laboratory of Image Science and Technology The Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University Nanjing210096 China The CAS Key Laboratory of FAST National Astronomical Observatory Chinese Academy of Sciences Beijing100101 China The Department of School of Computer Science and Technology Nanjing Tech University Jiangsu211816 China

Attributed graph clustering is one of the most fundamental tasks among graph learning field, the goal of which is to group nodes with similar representations into the same cluster without human annotations. Recent studies based on graph contrastive learning method have achieved remarkable results when exploit graph-structured data. However, most existing methods 1) do not directly address the clustering task, since the representation learning and clustering process are separated;2) depend too much on data augmentation, which greatly limits the capability of contrastive learning;3) ignore the contrastive message for clustering tasks, which adversely degenerate the clustering results. In this paper, we propose a Neighborhood Contrast Framework for Attributed Graph Clustering, namely NCAGC, seeking for conquering the aforementioned limitations. Specifically, by leveraging the Neighborhood Contrast Module, the representation of neighbor nodes will be'push closer' and become clustering-oriented with the neighborhood contrast loss. Moreover, a Contrastive Self-Expression Module is built by minimizing the node representation before and after the self-expression layer to constraint the learning of self-expression matrix. All the modules of NCAGC are optimized in a unified framework, so the learned node representation contains clustering-oriented messages. Extensive experimental results on four attributed graph datasets demonstrate the promising performance of NCAGC compared with 16 state-of-the-art clustering methods. The code is available at https://***/wangtong627/NCAGC. Copyright © 2022, The Authors. All rights reserved.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

AIDE: Attack Inference Based on Heterogeneous Dependency Graphs with MITRE ATT&CK

AIDE: Attack Inference Based on Heterogeneous Dependency Gra...

引用

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

作者： Weidong Zhou Chunhe Xia Nan Feng Xinyi Pan Tianbo Wang Xiaojian Li Beijing Key Lab. of Network Technology Beihang University Beijing China Guangxi Collaborative Innovation Center of Multi-Source Information Integration and Intelligent Processing Guangxi Normal University Guilin China School of Software & Microelectronics Peking University Beijing China School of Cyber Science and Technology Beihang University Beijing China School of Computer Science and Engineering Guangxi Normal University Guilin China

ISBN: (数字)9798331506209

ISBN: (纸本)9798331506216

The MITRE ATT&CK Framework provides a rich and actionable repository of adversarial tactics, techniques, and procedures (TTPs). By leveraging the associations among these techniques, we can infer unobserved techniques based on those have been observed. Current machine learning inference methods based on small-scale data face the following problems: "logical dependencies missing" and "complex associations missing." The former means that current association methods rely solely on the statistical dependencies between attack techniques while ignoring the logical dependencies involved in executing those techniques. The latter indicates that current inference methods simplify the complex associations between technologies based on statistical assumptions, failing to adequately capture the long-range and intricate real-world relationships between techniques. To address these problems, we propose an Attack Inference model based on heterogeneous DEpendency graphs (AIDE). First, we extract three types of logical dependencies: CVE, Pattern, and Conditional, which allows us to construct a heterogeneous dependency graph among the techniques, thus addressing the "logical dependencies missing" problem. Second, we employ Relation Graph Convolution Neural networks (RGCN) to aggregate heterogeneous dependencies into representation vectors of the techniques. Finally, we utilize the sequence inference model Decoder, which excels at capturing implicit contextual associations, to infer unknown attacks, thus addressing the "complex associations missing" problem. Experiments conducted on real datasets demonstrate that our model significantly outperforms state-of-the-art baseline models in both hits@k and map@k metrics.

关键词： Adaptation models Aggregates Refining Machine learning Vectors Threat assessment Real-time systems Decoding Security Context modeling

来源：评论

学校读者我要写书评

暂无评论

Asynchronous Tag-based Physical-Layer Authentication in Wireless Communications

引用

IEEE Transactions on Wireless Communications 2025年

作者： Tan, Haijun Xu, Yuntao Du, Jiewei Xie, Ning Li, Hongbin Peng Cheng Laboratory Control and Network Foundation Laboratory Frontier Research Center Shenzhen518107 China Shenzhen University State Key Laboratory of Radio Frequency Heterogeneous Integration The Guangdong Key Laboratory of Intelligent Information Processing College of Electronics and Information Engineering Shenzhen518060 China Beijing Jiaotong University School of Electronic and Information Engineering Beijing100044 China National Innovation Center of High Speed Train Qingdao266111 China Stevens Institute of Technology Department of Electrical and Computer Engineering HobokenNJ07030 United States

Tag-based physical-layer authentication (PLA) has gained significant research interest due to its high security and low complexity compared to traditional upper-layer authentication mechanisms. However, conventional tag-based PLA schemes often require strict synchronization between the tag and the source message before transmission, which may limit their practicality and effectiveness. To address this limitation, we propose an asynchronous tag-based PLA scheme, where the tag and the source message are transmitted asynchronously with an intentional time delay introduced at the transmitter. The proposed scheme enhances both compatibility and security without compromising robustness. We provide a theoretical analysis of the proposed scheme, and derive closed-form expressions to characterize its performance. Furthermore, theoretical comparisons with existing methods yield insightful conclusions regarding the advantages of the proposed approach. The scheme is implemented and extensively evaluated through simulations, where the theoretical results exhibit strong agreement with the simulation outcomes, thereby validating its effectiveness. © 2025 IEEE.

关键词： asynchronous closed form intentional timing delay Physical-layer authentication tag-based

来源：评论

学校读者我要写书评

暂无评论

Slnspeech: Solving Extended Speech Separation Problem with the Help of Sign Language

SSRN

引用

SSRN 2023年

作者： Wu, Jiasong Li, Xuan Li, Taotao Meng, Fanman Kong, Youyong Yang, Guanyu Senhadji, Lotfi Shu, Huazhong Laboratory of Image Science and Technology Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing210096 China Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing Southeast University Nanjing210096 China Univ Rennes INSERM LTSI - UMR 1099 RennesF-35000 France Univ Rennes INSERM RennesF-35000 France

Previous audio-visual speech separation methods use the synchronization of the speaker's facial movement and speech in the video to supervise the speech separation in a self-supervised way. In this paper, we propose a model to solve the speech separation problem assisted by both face and sign language, which we call the extended speech separation problem. We design a general deep learning network for learning the combination of three modalities, audio, face, and sign language information, for better solving the speech separation problem. To train the model, we introduce a large-scale dataset named the Sign Language News Speech (SLNSpeech) dataset in which three modalities of audio, face, and sign language coexist. Experiment results show that the proposed model has better performance and robustness than the usual audio-visual system. Besides, sign language modality can also be used alone to supervise speech separation task, and the introduction of sign language is helpful for hearing-impaired people to learn and communicate. Last, our model is a general speech separation framework and can achieve very competitive separation performance on two open-source audio-visual datasets. The code is available at https://***/iveveive/SLNSpeech. © 2023, The Authors. All rights reserved.

关键词： Audition

来源：评论

学校读者我要写书评

暂无评论

MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency Regularization

arXiv

引用

arXiv 2022年

作者： Duan, Yue Zhao, Zhen Qi, Lei Wang, Lei Zhou, Luping Shi, Yinghuan Gao, Yang The National Key Laboratory for Novel Software Technology The National Institute of Healthcare Data Science Nanjing University Nanjing210023 China The School of Electrical and Information Engineering The University of Sydney SydneyNSW2006 Australia The School of Computer Science and Engineering The Key Lab of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China The School of Computing and Information Technology University of Wollongong WollongongNSW2522 Australia

The core issue in semi-supervised learning (SSL) lies in how to effectively leverage unlabeled data, whereas most existing methods tend to put a great emphasis on the utilization of high-confidence samples yet seldom fully explore the usage of low-confidence samples. In this paper, we aim to utilize low-confidence samples in a novel way with our proposed mutex-based consistency regularization, namely MutexMatch. Specifically, the high-confidence samples are required to exactly predict "what it is" by conventional True-Positive Classifier, while low-confidence samples are employed to achieve a simpler goal — to predict with ease "what it is not" by True-Negative Classifier. In this sense, we not only mitigate the pseudo-labeling errors but also make full use of the low-confidence unlabeled data by consistency of dissimilarity degree. MutexMatch achieves superior performance on multiple benchmark datasets, i.e., CIFAR-10, CIFAR-100, SVHN, STL-10, mini-ImageNet and Tiny-ImageNet. More importantly, our method further shows superiority when the amount of labeled data is scarce, e.g., 92.23% accuracy with only 20 labeled data on CIFAR-10. Our code and model weights have been released at https://***/NJUyued/MutexMatch4SSL. Copyright © 2022, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Enhancing Interest Forwarding for Fast Recovery from Unanticipated Data Access Failure in NDN

引用

China Communications 2019年第7期16卷 120-130页

作者： Xiaoyan Hu Xuhui Liu Lixia Zhao Jian Gong Guang Cheng School of Cyber Science & Engineering Southeast University Nanjing 211189 China Key Laboratory of Computer Network and Information Integration of Ministry of Education Southeast University Nanjing China Key Laboratory of Computer Network Technology of Jiangsu Province Nanjing China

We show that an aggregated Interest in Named Data networking (NDN) may fail to retrieve desired data since the Interest previously sent upstream for the same content is judged as a duplicate one and then dropped by an upstream node due to its multipath forwarding. Furthermore, we propose NDRUDAF, a NACK based mechanism that enhances the Interest forwarding and enables Detection and fast Recovery from such Unanticipated Data Access Failure. In the NDN enhanced with NDRUDAF, the router that aggregates the Interest detects such unanticipated data access failure based on a negative acknowledgement from the upstream node that judges the Interest as a duplicate one. Then the router retransmits the Interest as soon as possible on behalf of the requester whose Interest is aggregated to fast recover from the data access failure. We qualitatively and quantitatively analyze the performance of the NDN enhanced with our proposed NDRUDAF and compare it with that of the present NDN. Our experimental results validate that NDRUDAF improves the system performance in case of such unanticipated data access failure in terms of data access delay and network resource utilization efficiency at routers.

关键词： named data networking interest aggregation multipath forwarding data access failure negative acknowledgement

来源：评论

学校读者我要写书评

暂无评论

IP Traffic Behavior Characterization via Semantic Mining

SSRN

引用

SSRN 2022年

作者： Zang, Xiaodong Gong, Jian Gao, Peng Zhang, Guowei School of Cyber Science and Engineering Qufu Normal University Qufu273100 China School of Cyber Science and Engineering Southeast University Nanjing211189 China Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China

Discovering and describing IP traffic behavior is becoming more and more significant for efficient network management and security monitoring. Although the vast majority of modeling techniques have been proposed, most of which focused on the properties based on a single dimension, such as volume-based or spatial-based, resulting in incomplete behavior analysis. In this paper, we demonstrate that IP traffic behavior can be profiled from the perspective of end users’ access relationship. Not only groups of IPs having similar behavior traits, but also malicious behavior can be found by profiling their behavior pattern in the overwhelming measurement data. To this end, firstly, a collection of features are derived, which can profile the traffic behavior from four dimensions, such as temporal, spatial, category, and intensity. Then, the rhythmic behavior, the cyclical behavior, the access stable behavior, the service diversity behavior, and the hotspot behavior are characterized and modeled. Finally, we use synthetic data and the real Internet Netflow data collected from the China Education Research network backbone (CERNET) to empirically validate our proposal. Extensive results demonstrate that the applications of derived traffic patterns can achieve fine-grained traffic monitoring. © 2022, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

BRFL: A Blockchain-based Byzantine-Robust Federated Learning Model

arXiv

引用

arXiv 2023年

作者： Li, Yang Xia, Chunhe Li, Chang Wang, Tianbo The Key Laboratory of Beijing Network Technology Beihang University Beijing100191 China The Guangxi Collaborative Innovation Center of Multi-Source Information Integration and Intelligent Processing Guangxi Normal University Guilin541004 China The School of Computer Science and Technology Zhengzhou University of Light Industry Zhengzhou450000 China The School of Cyber Science and Technology Beihang University Beijing100191 China The Shanghai Key Laboratory of Computer Software Evaluating and Testing Shanghai201112 China

With the increasing importance of machine learning, the privacy and security of training data have become critical. Federated learning, which stores data in distributed nodes and shares only model parameters, has gained significant attention for addressing this concern. However, a challenge arises in federated learning due to the Byzantine Attack Problem, where malicious local models can compromise the global model's performance during aggregation. This article proposes the Blockchainbased Byzantine-Robust Federated Learning (BRLF) model that combines federated learning with blockchain technology. This integration enables traceability of malicious models and provides incentives for locally trained clients. Our approach involves selecting the aggregation node based on Pearson's correlation coefficient, and we perform spectral clustering and calculate the average gradient within each cluster, validating its accuracy using local dataset of the aggregation nodes. Experimental results on public datasets demonstrate the superior byzantine robustness of our secure aggregation algorithm compared to other baseline byzantine robust aggregation methods, and proved our proposed model effectiveness in addressing the resource consumption problem. Copyright © 2023, The Authors. All rights reserved.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Rethinking calibration of deep neural networks: do not be afraid of overconfidence 21

Rethinking calibration of deep neural networks: do not be af...

引用

Proceedings of the 35th International Conference on Neural information Processing Systems

作者： Deng-Bao Wang Lei Feng Min-Ling Zhang School of Computer Science and Engineering Southeast University Nanjing China and Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education China College of Computer Science Chongqing University Chongqing China

ISBN: (纸本)9781713845393

Capturing accurate uncertainty quantification of the predictions from deep neural networks is important in many real-world decision-making applications. A reliable predictor is expected to be accurate when it is confident about its predictions and indicate high uncertainty when it is likely to be inaccurate. However, modern neural networks have been found to be poorly calibrated, primarily in the direction of overconfidence. In recent years, there is a surge of research on model calibration by leveraging implicit or explicit regularization techniques during training, which achieve well calibration performance by avoiding overconfident outputs. In our study, we empirically found that despite the predictions obtained from these regularized models are better calibrated, they suffer from not being as calibratable, namely, it is harder to further calibrate these predictions with post-hoc calibration methods like temperature scaling and histogram binning. We conduct a series of empirical studies showing that overconfidence may not hurt final calibration performance if post-hoc calibration is allowed, rather, the penalty of confident outputs will compress the room of potential improvement in post-hoc calibration phase. Our experimental findings point out a new direction to improve calibration of DNNs by considering main training and post-hoc calibration as a unified framework.

关键词：

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：