检索结果-内蒙古大学图书馆

The Journal of Machine Learning Research 2024年第1期25卷 4098-4180页

作者： Zhen Fang Yixuan Li Feng Liu Bo Han Jie Lu Australian Artificial Intelligence Institute University of Technology Sydney Ultimo NSW Australia Department of Computer Sciences The University of Wisconsin Madison Madison WI School of Computing and Information Systems The University of Melbourne Carlton VIC Australia Department of Computer Science Hong Kong Baptist University Kowloon Tong Hong Kong SAR

Supervised learning aims to train a classifier under the assumption that training and test data are from the same distribution. To ease the above assumption, researchers have studied a more realistic setting: out-of-distribution (OOD) detection, where test data may come from classes that are unknown during training (i.e., OOD data). Due to the unavailability and diversity of OOD data, good generalization ability is crucial for effective OOD detection algorithms, and corresponding learning theory is still an open problem. To study the generalization of OOD detection, this paper investigates the probably approximately correct (PAC) learning theory of OOD detection that fits the commonly used evaluation metrics in the literature. First, we find a necessary condition for the learnability of OOD detection. Then, using this condition, we prove several impossibility theorems for the learnability of OOD detection under some scenarios. Although the impossibility theorems are frustrating, we find that some conditions of these impossibility theorems may not hold in some practical scenarios. Based on this observation, we next give several necessary and sufficient conditions to characterize the learnability of OOD detection in some practical scenarios. Lastly, we offer theoretical support for representative OOD detection works based on our OOD theory.

关键词： out-of-distribution detection weakly supervised learning learnability

来源：评论

学校读者我要写书评

暂无评论

The future of E-commerce in the publishing industry 2

The future of E-commerce in the publishing industry

引用

2nd International Conference on Smart computing and Electronic Enterprise, ICSCEE 2021

作者： Aseh, Khairi Al-Qudah, Omar Mohammed Ali Abdullah Yusoff, M. Hafiz Pathmanathan, P. Ravindran Latiff, Nurul Adilah Abdul Jusoh, Julaily Aida Abu-Ulbeh, Waheeb El-Ebiary, Yousef A. Baker Al Moaiad, Yazeed International Institute of Applied Science Swiss School of Management Netherlands Irbid National University Faculty of Financial and Business Science Head of E-Marketing Department Jordan Albukhary International University School of Computing and Informatic Kedah Malaysia Faculty of Informatics and Computing UniSZA Malaysia UMT Faculty of Ocean Engineering Technology and Informatics Malaysia UniSZA Faculty of Informatics and Computing Malaysia Faculty of Computer Science and Informatics Cybersecurity Department Jordan MEDIU Faculty of Computer and Information Technology Malaysia

ISBN: (纸本)9781665432221

In comparison to conventional book-printed content, electronic "publishing"has allowed users to import files to either a personal computer or an e-book reader to read text in a digital format. Reduced printing costs and reduced time to market by streamlined drafting, proofing, and printing methods are the benefits of the publishing industry over the conventional printed book (p-book). It prevents binding, distribution and shipping. The intent of this paper is to explore the strategic use of e-commerce in the transformation of the publishing industry. The first part of the paper discusses the strategic use of e-commerce/Web-Based Business in the transition from a single distribution channel mass producer of printed books to a multiple distribution channel that will include e-books and print on demand technology in the publishing industry. Emphasis is put on management techniques in the global marketplace for the use of emerging information technologies. The second section of this paper addresses the transfer in competitive influence between the different levels of the supply chain of the publishing industry, including the latest trends in electronic publishing from the stakeholders' point of view. Expected to serve as the base for development within the digitized print industry. As e-publishing becomes popular and moves towards maturity, to allow a "mainstream"status, common standards must emerge and combine. © 2021 IEEE.

关键词： Electronic publishing

来源：评论

学校读者我要写书评

暂无评论

IoT Security Frameworks: A Comparative Review with a Focus on Privacy

IoT Security Frameworks: A Comparative Review with a Focus o...

引用

Cyber Resilience (ICCR), International Conference on

作者： Abduljaleel Al-Hasnawi Yitong Niu Jamal F. Tawfeq Manas Ranjan Pradhan Mohammed Salahat Taher M. Ghazal Ministry of Higher Education and Scientific Research Baghdad Iraq Department of Communication Technology Engineering College of Information Technology Imam Ja'afar Al-Sadiq University Baghdad Iraq School of Aeronautical Engineering AnYang University Anyang China Department of Medical Instrumentation Technical Engineering Medical Technical College Al-Farahidi University Baghdad Iraq School of Computing Skyline University College University City Sharjah Sharjah UAE College of Engineering and Technology University of Science and Technology of Fujairah Fujairah UAE Computer Science Department Centre for Cyber Physical Systems Khalifa University Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia (UKM) Bangi Selangor Malaysia

ISBN: (数字)9798350394962

ISBN: (纸本)9798350394979

The privacy of sensitive data within the Internet of Things (IoT) is at risk due to potential privacy violations. These violations can occur through illegal access by external entities, as well as inadvertent disclosures or intentional attacks by those within the system who exceed their authorized access privileges. Furthermore, there needs to be more control experienced by data owners regarding their sensitive data as it transitions from their domain to other entities within the Internet of Things (IoT) ecosystem. Numerous methods have been presented in the academic literature over the past few years to effectively tackle the privacy concerns associated with the Internet of Things (IoT). This paper aims to classify existing literature by organizing it according to the specific technology or technique employed to safeguard privacy. The classification encompasses four primary areas of privacy solutions, specifically, solutions based on encryption, solutions based on identity management, solutions based on policy enforcement, and solutions focused on self-protection. The paper examines various state-of-the-art solutions within each category. It begins by providing a broad comparison between these solutions and our work. Subsequently, it focuses on a specific comparison of the privacy objectives of the offered PEFM solutions.

关键词： Privacy Data privacy Reviews Ecosystems Encryption Internet of Things Resilience

来源：评论

学校读者我要写书评

暂无评论

Graph Learning-Based Multi-User Multi-Task Offloading in Wireless computing Power Networks

引用

IEEE Internet of Things Journal 2025年

作者： Dai, Yueyue Rao, Xiaoyang Gu, Bruce Qu, Youyang Yang, Huiran Lu, Yunlong Huazhong University of Science and Technology Research Center of6G Mobile Communications School of Cyber Science and Engineering Wuhan430074 China Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education China Australia University of Chinese Academy of Sciences Institute of Information Engineering Chinese Academy of Sciences Beijing China and the School of Cyber Security Beijing China Beijing Jiaotong University School of Electronic and Information Engineering Beijing100044 China

To enhance service quality, wireless computing power networks (WCPNs) need to realize flexible scheduling and allocation of computation resources across heterogeneous computing servers. Due to large user scales and diverse computation tasks, it is difficult for the current WCPN to serve multiple users and handle multiple tasks concurrently. Graph learning is a promising approach that can learn the representations of nodes through graph structures, enabling the exploration of dependencies among multiple users and tasks, and thereby facilitating computation task offloading. In this paper, we propose a graph learning-based multi-user multi-task offloading scheme for WCPN. First, we propose a wireless computing power network with multi-user and multi-task in which users need to make full use of distributed computing resources through task offloading to ensure efficient task execution. We formulate a system energy consumption minimization problem to jointly optimize computation resources, transmission power, and task offloading. To address the problem, we utilize graph learning to transform the joint optimization problem into a graph regression problem and leverage line graph to explore the solution. Numerical results demonstrate that our proposed scheme can improve computation efficiency, enhance optimization performance, and maintain transferability compared with the benchmarks. © 2014 IEEE.

关键词： computing power

来源：评论

学校读者我要写书评

暂无评论

Incomplete Multi-view Clustering Based on Joint Concept Decomposition and Anchor Graph Learning 11th

Incomplete Multi-view Clustering Based on Joint Concept Dec...

引用

11th International Joint Conference on Rough Sets, IJCRS 2025

作者： Li, Zhuowen Chen, Hongmei Xiang, Biao Yuan, Zhong Luo, Chuan Li, Tianrui School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China National Engineering Laboratory of Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu611756 China Engineering Research Center of Sustainable Urban Intelligent Transportation Ministry of Education Chengdu611756 China Manufacturing Industry Chains Collaboration and Information Support Technology Key Laboratory of Sichuan Province Southwest Jiaotong University Chengdu611756 China College of Computer Science Sichuan University Chengdu610065 China

ISBN: (纸本)9783031927430

The main objective of incomplete multi-view clustering is to effectively utilize the existing view information to fill in the missing data and to mine the complementary information and potential associations between multiple views to effectively group samples. The existing primary means for recovering missing information are classified into two parts: one uses matrix decomposition or low-rank constraints to fill in the missing views, and the other constructs multiple graph structures and uses graph regularization to recover the corresponding missing parts. However, the two methods are relatively independent and are not used jointly to improve the model performance. This paper proposes an incomplete multi-view clustering method with complete clustering metrics and complete anchor graphs (CFAG). Specifically, each view’s missing samples are recovered by a matrix of consistent clustering metrics and represented by an anchor graph. The recovered complete data performs anchor graph construction and anchor learning and utilizes the orthogonal variation principle to learn the consistency structure of multiple anchor graphs to mine the complementary information among multiple views. The clustering metrics are obtained based on the consistency structure, and the consistency of the clustering metrics is also considered to obtain the final clustering results. The effectiveness of the proposed method is verified on multiple datasets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Rainfall Induced Landslide Detection in Study Area using Machine Learning Model

Rainfall Induced Landslide Detection in Study Area using Mac...

引用

International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)

作者： Pradeep Singh Rawat Anuj Kumar Yadav Varun Barthwal Sachin Gaur School of Computing DIT University Dehradun Dehradun India Department of Information Technology H.N.B Garhwal University Pauri Pauri Garhwal India Department of Computer Science and Engineering B.T Kumaon Institute of Technology Dwarahat Dwarahat India

In the present era of computing, communication and technology, natural disaster can be controlled in an efficient manner. The natural disaster in hill sate is common. It can be mitigated using machine learning techniques. In this manuscript out objective is proposed a machine learning model which focus on rainfall induced landslide prediction in uttarakhand state districts using benchmark dataset. There is a good correlation between landslide and antecedent rain fall. The antecedent rain fall supports the machine learning model for better accuracy and correctness. The machine learning model with optimal performance metrics provides the prior information about the level rainfall and its impact level on landslide in a study area of the focused state. The results show that Random Forest model outperforms the linear model, SVR model, and neural network model respectively. The key performance indicators i.e. mean absolute error(MAE), and root mean square error(RMSE) are improved by a factor of 79.05%, and 83.34% respectively. The key performance indicators evaluated and analyzed against state of art methodologies i.e. Random Forest model outperforms the linear model, SVR model, and neural network

关键词：

来源：评论

学校读者我要写书评

暂无评论

Federated Causally Invariant Feature Learning 39

Federated Causally Invariant Feature Learning

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Guo, Xianjie Yu, Kui Cui, Lizhen Yu, Han Li, Xiaoxiao School of Computer Science and Information Engineering Hefei University of Technology China Key Laboratory of Knowledge Engineering with Big Data of Ministry of Education China School of Software Shandong University China College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering The University of British Columbia Canada Vector Institute Canada

ISBN: (纸本)157735897X

Federated feature selection (FFS) is a promising field for selecting informative features while preserving data privacy in federated learning (FL) settings. Existing FFS methods focus on capturing the correlations between features and labels. They struggle to achieve satisfactory performance in the face of data distribution heterogeneity among FL clients, and cannot address the out-of-distribution (OOD) problem that arises when a significant portion of clients do not actively participate in FL training. To address these limitations, we propose Federated Causally Invariant Feature Learning (FedCIFL), a novel approach for learning causally invariant features in a privacy-preserving manner. We design a sample reweighting strategy to eliminate spurious correlations introduced by selection bias and iteratively estimate the federated causal effect between each feature and the labels (with the remaining features initially treated as confounders). By iteratively refining the confounding feature set to identify the true confounders, FedCIFL mitigates the impact of limited local data on the accuracy of federated causal effect estimation. Theoretical analysis proves the correctness of FedCIFL under reasonable assumptions. Extensive experiments on synthetic and real-world datasets demonstrate the superiority of FedCIFL against eight state-of-the-art baselines, beating the best-performing approach by 3.19%, 9.07% and 2.65% in terms of average test Accuracy, RMSE and F1 score, respectively. It is a first-of-its-kind FFS approach capable of handling Non-IID and OOD data simultaneously. The source code is available at https://***/Xianjie-Guo/FedCIFL. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Using Pattern Analysis and Machine Learning to Categorise users of Online Directories based on their Surfing Habits

Using Pattern Analysis and Machine Learning to Categorise us...

引用

Sustainable Communication Networks and Application (ICSCNA), International Conference on

作者： S. Athinarayanan M. Robinson Joel Shaik Jumlesha K. Susmitha Limgamgunta Charitha Department of CSE School of Computing Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Avadi Chennai Tamil Nadu India Department of Computer Science and Engineering AITS Tirupati Andra Pradesh Department of Information Technology Kings Engineering College Chennai India

The major goal is to group users together, including in student organizations. Assume that the notion is required to cluster consumers in educational institutions. A cluster of websites or web pages organized by user interest makes up an online use directory. Web mining is the process of using strategies to extract websites from the web proxy log file. We are employing the Apriori Algorithm and the Fuzzy Clustering technique for it. Typically, data cleaning, a preprocessing step, is followed by aggregation of the data after cleanliness in web use mining. We do a pattern analysis following the data's classification. The outcomes of the studies are contrasted with another intriguing technique that is made public. In comparison to other efforts, the outcomes produced by the suggested system exhibit better performance. The technique we utilised, called WDUM or web directories user mining, produces better results than earlier techniques. We compare user growth and penetration performance to the threshold values. Thus, after preprocessing, the log files are retrieved, and the Apriori Algorithm and fuzzy logic classification are applied. These outcomes demonstrate the greatest methods for capturing user interest. Finally, the projected user clicked the prediction findings using machine learning techniques.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound

arXiv

引用

arXiv 2023年

作者： Cai, Hanbo Zhang, Pengcheng Dong, Hai Xiao, Yan Koffas, Stefanos Li, Yiming College of Computer and Information Hohai University Nanjing China School of Computing Technologies RMIT University Melbourne Australia School of Cyber Science and Technology Sun Yat-sen University Shenzhen China Delft University of Technology Delft Netherlands Tsinghua University Shenzhen China Ant Group Hangzhou China

Deep neural networks (DNNs) have been widely and successfully adopted and deployed in various applications of speech recognition. Recently, a few works revealed that these models are vulnerable to backdoor attacks, where the adversaries can implant malicious prediction behaviors into victim models by poisoning their training process. In this paper, we revisit poison-only backdoor attacks against speech recognition. We reveal that existing methods are not stealthy since their trigger patterns are perceptible to humans or machine detection. This limitation is mostly because their trigger patterns are simple noises or separable and distinctive clips. Motivated by these findings, we propose to exploit elements of sound (e.g., pitch and timbre) to design more stealthy yet effective poison-only backdoor attacks. Specifically, we insert a short-duration high-pitched signal as the trigger and increase the pitch of remaining audio clips to ‘mask’ it for designing stealthy pitch-based triggers. We manipulate timbre features of victim audios to design the stealthy timbre-based attack and design a voiceprint selection module to facilitate the multi-backdoor attack. Our attacks can generate more ‘natural’ poisoned samples and therefore are more stealthy. Extensive experiments are conducted on benchmark datasets, which verify the effectiveness of our attacks under different settings (e.g., all-to-one, all-to-all, clean-label, physical, and multi-backdoor settings) and their stealthiness. The code for reproducing main experiments are available at https://***/HanboCai/BadSpeech_SoE. © 2023, CC BY.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Enabling Quadruple-D Compensation Coil Integration for Efficient Power Transfer to Receiver in Wireless Power Transfer Systems for EV Charging

引用

IEEE Journal of Emerging and Selected Topics in Power Electronics 2025年

作者： Erel, Mehmet Zahid Imtiaz, Tabish Elsanabary, Ahmed Mubin, Marizan Soon, Tey Kok Mekhilef, Saad Ankara Yildirim Beyazit University Department of Energy Systems Engineering Ankara Turkey Universiti Malaya Faculty of Engineering Department of Electrical Engineering Kuala Lumpur Malaysia Port Said University Faculty of Engineering Department of Electrical Engineering Port Said Egypt Universiti Malaya Department of Computer System and Technology Faculty of Computer Science and Information Technology Kuala Lumpur Malaysia Swinburne University of Technology School of Science Computing and Engineering Technologies Hawthorn Australia Presidency University Department of Electrical and Electronics Engineering Karnataka Bengaluru India

Wireless power transfer (WPT) presents a promising approach for charging electric vehicles (EVs), offering increased convenience, reliability, and safety for EV customers. Recently, the incorporation of multiple coil structures in wireless charging systems has addressed various operational challenges, enhanced power transfer, and provided a scalable solution for EVs. However, cross-coupling issues and space limitation problems arise when multiple coils are introduced to the WPT system. In this paper, a novel integrated compact coupler is proposed, including two naturally decoupled coils on the transmitter side, while both are coupled with the receiver coil to improve the power transfer capability of the system. ANSYS Maxwell is employed to simulate the proposed coupler and evaluate its performance under misalignment operations. The proposed wireless charging system inherently detects misalignment direction based on the variation of output voltage with respect to x-misalignment. As a result, the experimental prototype of the designed wireless WPT system presents a peak efficiency of 97.08% at 1.147-kW output power for an air gap of 120 mm under fully-aligned condition. Additionally, the designed compensation coil can transfer power of 243.76-W under fully-aligned condition and increases reliability in power transfer during misalignment operations. © 2013 IEEE.

关键词： Inductive power transmission

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：