检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Yi, Liping Yu, Han Ren, Chao Wang, Gang Liu, Xiaoguang Li, Xiaoxiao College of Computer Science TMCC SysNet DISSec GTIISC Nankai 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

Model heterogeneous federated learning (MHeteroFL) enables FL clients to collaboratively train models with heterogeneous structures in a distributed fashion. However, existing MHeteroFL methods rely on training loss to transfer knowledge between the client model and the server model, resulting in limited knowledge exchange. To address this limitation, we propose the Federated model heterogeneous Matryoshka Representation Learning (FedMRL) approach for supervised learning tasks. It adds an auxiliary small homogeneous model shared by clients with heterogeneous local models. (1) The generalized and personalized representations extracted by the two models' feature extractors are fused by a personalized lightweight representation projector. This step enables representation fusion to adapt to local data distribution. (2) The fused representation is then used to construct Matryoshka representations with multi-dimensional and multi-granular embedded representations learned by the global homogeneous model header and the local heterogeneous model header. This step facilitates multi-perspective representation learning and improves model learning capability. Theoretical analysis shows that FedMRL achieves a O(1/T) non-convex convergence rate. Extensive experiments on benchmark datasets demonstrate its superior model accuracy with low communication and computational costs compared to seven state-of-the-art baselines. It achieves up to 8.48% and 24.94% accuracy improvement compared with the state-of-the-art and the best same-category baseline, respectively. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Inductive Relation Prediction by Disentangled Subgraph Structure

引用

Tsinghua science and Technology 2024年第5期29卷 1566-1579页

作者： Guiduo Duan Rui Guo Wenlong Luo Guangchun Luo Tianxi Huang Laboratory of Intelligent Collaborative Computing University of Electronic Science and Technology of ChinaChengdu 611731Chinaand also with Trusted Cloud Computing and Big Data Key Laboratory of Sichuan ProvinceChengdu 611731China School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China Hangzhou NetEase Cloud Music Technology Co. Ltd.Hangzhou 310052China Department of Fundamental Courses Chengdu Textile CollegeChengdu 611700China

Currently,most existing inductive relation prediction approaches are based on subgraph structures,with subgraph features extracted using graph neural networks to predict ***,subgraphs may contain disconnected regions,which usually represent different semantic *** not all semantic information about the regions is helpful in relation prediction,we propose a relation prediction model based on a disentangled subgraph structure and implement a feature updating approach based on relevant semantic *** indirectly achieve the disentangled subgraph structure from a semantic perspective,the mapping of entity features into different semantic spaces and the aggregation of related semantics on each semantic space are *** disentangled model can focus on features having higher semantic relevance in the prediction,thus addressing a problem with existing approaches,which ignore the semantic differences in different subgraph ***,using a gated recurrent neural network,this model enhances the features of entities by sorting them by distance and extracting the path information in the ***,it is shown that when there are numerous disconnected regions in the subgraph,our model outperforms existing mainstream models in terms of both Area Under the Curve-Precision-Recall(AUC-PR)and Hits@*** prove that semantic differences in the knowledge graph can be effectively distinguished and verify the effectiveness of this method.

关键词： disentangled subgraph structure knowledge graph completion Gated Recurrent Unit(GRU) feature updating

来源：评论

学校读者我要写书评

暂无评论

A Deep Reinforcement Learning-Based Task Offloading Framework for Edge-Cloud Computing 2

A Deep Reinforcement Learning-Based Task Offloading Framewor...

引用

2nd IEEE International Conference on Integrated Circuits and Communication Systems, ICICACS 2024

作者： Kalpana, Ponugoti Almusawi, Muntather Chanti, Yerrolla Sunil Kumar, V. Varaprasad Rao, M. Avn Institute of Engineering & Technology Department of Computer Science and Engineering Telangana India The Islamic University Najaf Iraq Sr University School of Computer Science & Artificial Intelligence Warangal India Nitte Meenakshi Institute of Technology Department of Ai and Ml Bengaluru India Cvr College of Engineering Department of Data Science Hyderabad India

ISBN: (纸本)9798350317558

By introducing the processing and resources storage to the network edge, edge computing is a new, promising computing paradigm that drastically decreases network traffic and service latency. Many edge computing applications consist of interdependent tasks, wherein the results of one task are the inputs of another. The important and difficult problem of where to position each running task to optimise Quality-of-Service (QoS) is how to offload these tasks to edge of network. In this work, implemented a novel Deep Reinforcement Learning based Task Offloading (DRLTO) method utilized as the intelligent task offloading that uses a Directed Acyclic Graph (DAG) to represent the dependent tasks and off-policy reinforcement learning powered by a Sequence-to-Sequence (S2S) neural network. This research outcomes show that the DRLTO achieved less cloud processing time, and number of single terminal tasks, and percentage of failed task when compared to Deep reinforcement learning-based cloud-edge collaborative mobile computation offloading (DRL-CCMCO) and DeepEdge. © 2024 IEEE.

关键词： and task offloading Deep reinforcement learning Multi-access edge computing sequence to sequence neural networks

来源：评论

学校读者我要写书评

暂无评论

Gradient-Free Methods for Nonconvex Nonsmooth Stochastic Compositional Optimization 38

Gradient-Free Methods for Nonconvex Nonsmooth Stochastic Com...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Liu, Zhuanghua Luo, Luo Low, Bryan Kian Hsiang Department of Computer Science National University of Singapore CNRS@CREATE Ltd. 1 Create Way #08-01 CREATE Tower 138602 Singapore School of Data Science Fudan University Shanghai Key Laboratory for Contemporary Applied Mathematics China Department of Computer Science National University of Singapore Singapore

Stochastic compositional optimization (SCO) problems are popular in many real-world applications, including risk management, reinforcement learning, and meta-learning. However, most of the previous methods for SCO require the smoothness assumption on both the outer and inner functions, which limits their applications to a wider range of problems. In this paper, we study the SCO problem in that both the outer and inner functions are Lipschitz continuous but possibly nonconvex and nonsmooth. In particular, we propose gradient-free stochastic methods for finding the (δ, ϵ)-Goldstein stationary points of such problems with non-asymptotic convergence rates. Our results also lead to an improved convergence rate for the convex nonsmooth SCO problem. Furthermore, we conduct numerical experiments to demonstrate the effectiveness of the proposed methods. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Developments and Applications of data Deidentification Technology under Big data

引用

Journal of Electronic science and Technology 2017年第3期15卷 231-239页

作者： Hung-Li Chen Yao-Tung Tsou Bo-Chen Tai Szu-Chuang Li Yen-Nun Huang Chia-Mu Yu Yu-Shian Chiu Department of Communications Engineering Feng Chia University Taichung 400 Research Center for Information Technology Innovation Academia Sinica Taipei 100 Department of Computer Science and Engineering National Chung Hsing University Data Analytics Technology &Applications Institute for Information Industry Taipei 100

In this age characterized by rapid growth in the volume of data,data deidentification technologies have become crucial in facilitating the analysis of sensitive *** instance,healthcare information must be processed through deidentification procedures before being passed to data analysis agencies in order to prevent any exposure of personal details that would violate *** such,privacy protection issues associated with the release of data and data mining have become a popular field of study in the domain of big *** a strict and verifiable definition of privacy,differential privacy has attracted noteworthy attention and widespread research in recent *** this study,we analyze the advantages of differential privacy protection mechanisms in comparison to traditional deidentification data protection ***,we examine and analyze the basic theories of differential privacy and relevant studies regarding data release and data mining.

关键词： Index Terms--Deidentification differential privacy.

来源：评论

学校读者我要写书评

暂无评论

Hybrid Framework for Sentiment Analysis Using ConvBiLSTM and BERT 2

Hybrid Framework for Sentiment Analysis Using ConvBiLSTM and...

引用

2nd International Conference on Technological Advancements in Computational sciences, ICTACS 2022

作者： Bhola, Abhishek Athithan, Senthil Singh, Shashank Mittal, Sandeep Sharma, Yogesh Kumar Dhatterwal, Jagjit Singh Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering Andhra Pradesh Vaddeswaram India Thapar Institute of Engineering and Technology Department of CSE Patiala India Rd Engineering College Department of Computer Science and Engineering Ghaziabad India Koneru Lakshmaiah Education Foundation Department of Artificial Intelligence & Data Science Andhra Pradesh Vaddeswaram India

ISBN: (数字)9781665476577

ISBN: (纸本)9781665476577

Sentiment analysis is specifically a text mining technique that utilizes natural language processing to computerize the process of analyzing text that aims to determine the sentiment expressed. The fundamental purpose of sentiment analysis is to get valuable insights that lead to all-around development in specific domains. The fantastic applications of sentimental analysis include monitoring social media, management of customer support, and customer reviews research. One of the major pitfalls in sentiment analysis is word ambiguity. To overcome this drawback, a proposed hybrid framework presented in this work is capable of dealing with such ambiguity issues. The considered evaluation parameters are accuracy, F1 score and time taken. The proposed hybrid framework utilizes Convolutional Bi-directional Long short-term memory network (ConvBiLSTM) with Bidirectional Encoder representations from Transformer (BERT) tokeniser on the given dataset and outperform other methodologies with 95.10% accuracy. © 2022 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Automatic Early Detection of Tomato Blight Disease Using Deep Neural Networks 4

Automatic Early Detection of Tomato Blight Disease Using Dee...

引用

4th International Conference on Technological Advancements in Computational sciences, ICTACS 2024

作者： Halder, Anangsha Neogi, Subhrendu Guha Dutta, Shibdas Chandra, Soumallya Chandra Majumdar, Pritha Guha Sarkar, Akash The Neotia University Department of Computer Science and Engineering West Bengal Sarisha India Brainware University Department of Computer Science and Engineering West Bengal Kolkata India Dcg Data-Core System India Pvt. Ltd West Bengal Kolkata India

ISBN: (纸本)9798350387490

Crop disease control is crucial for food security since plant illnesses diminish agricultural production. Crop productivity and quality depend on leaf diseases like tomatoes. Early plant disease identification utilizing computer vision and artificial intelligence can lessen disease severity and overcome the limitations of continuous human monitoring. Automatic Early detection and alert systems reduce disease and boost agricultural productivity by classifying tomato diseases from tomato leaf images using Deep Neural Network Models. Our study examined leaf photos to classify tomato diseases and healthy plants using AlexNet, ResNet-9, VGG-19, and a modified CNN model for tomato plant disease diagnosis using deep neural network architectures. The proposed CNN analyses images to identify edges, colours, and textures. The suggested model can recognize healthy and diseased tomato leaves with 98% accuracy, making it a valuable crop management tool that promotes sustainable agriculture and global food security. © 2024 IEEE.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Credit Card Batch Processing in Banking System 2nd

Credit Card Batch Processing in Banking System

引用

2nd International Conference on Advances in Computing Research, ACR 2024

作者： Poudel, Samir Movinuddin Gutta, Sanjana Kommu, Revanth Kumar Upadhyay, Jiblal Hasan, Md Nahid Poudel, Khem Department of Computer Science Middle Tennessee State University MurfreesboroTN37132 United States Computational and Data Science Middle Tennessee State University MurfreesboroTN37132 United States

ISBN: (纸本)9783031569494

Batch payments, involving a single transaction to multiple recipients via a single bank account, offer an efficient alternative to individual payments. This method condenses multiple payments into a single debit entry on bank statements, accelerating payment processing and enhancing business efficiency. Utilizing bank wire transfers is a prevalent means of initiating batch payments. In contrast to real-time processing, where transactions are handled immediately, batch processing entails aggregating authorized credit card transactions by merchants for submission to their credit card processors at the close of each business day or at scheduled intervals [1]. Merchant-authorized credit card transactions are compiled and sent to customers’ banks, seeking authorization. Upon approval, funds are transferred to the business’s bank account. Our objective is to identify fraudulent credit card transactions to safeguard credit card customers from erroneous charges for unauthorized purchases. The model employed aims to be both swift and adept at detecting anomalies, swiftly classifying potentially fraudulent transactions. The study involved the evaluation of multiple classifiers such as Gradient-Boosted Tree (GBT) and Random Forest, revealing that while the GBT classifier exhibited exceptional precision, the Random Forest classifier emerged as the preferred choice for our dataset. The selection was based on practical considerations, including efficiency and ease of implementation. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Handbook of Evolutionary Machine Learning 1

引用

丛书名： Genetic and Evolutionary Computation

1000年

作者： Wolfgang Banzhaf Penousal Machado Mengjie Zhang

ISBN: (数字)9789819938148

ISBN: (纸本)9789819938131;9789819938162

This book, written by leading international researchers of evolutionary approaches to machine learning, explores various ways evolution can address machine learning problems and improve current methods of machine learning. Topics in this book are organized into five parts. The first part introduces some fundamental concepts and overviews of evolutionary approaches to the three different classes of learning employed in machine learning. The second addresses the use of evolutionary computation as a machine learning technique describing methodologic improvements for evolutionary clustering, classification, regression, and ensemble learning. The third part explores the connection between evolution and neural networks, in particular the connection to deep learning, generative and adversarial models as well as the exciting potential of evolution with large language models. The fourth part focuses on the use of evolutionary computation for supporting machine learning methods. This includes methodological developments for evolutionary data preparation, model parametrization, design, and validation. The final part covers several chapters on applications in medicine, robotics, science, finance, and other disciplines. Readers find reviews of application areas and can discover large-scale, real-world applications of evolutionary machine learning to a variety of problem domains.;This book will serve as an essential reference for researchers, postgraduate students, practitioners in industry and all those interested in evolutionary approaches to machine learning.

关键词： Artificial Intelligence Machine Learning Computational Intelligence Evolutionary Biology

来源：评论

学校读者我要写书评

暂无评论

An Efficient Unique Tokenized Authentication Technique Based data Offloading Approach Using Key Exchange to Maintain data Integrity in Heterogeneous Networks

引用

SN computer science 2023年第5期4卷 1-9页

作者： Mandali, Ramesh Tirapathi Reddy, B. Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation (K.L.E.F) Green Fields Andhra Pradesh Vaddeswaram 522502 India Department of Artificial Intelligence and Data Science Koneru Lakshmaiah Education Foundation Vaddeswaram India

In the current period, the number of mobile users is rapidly expanding with a complicated network setup with a large amount of network traffic, which results in data overload. Insufficient network bandwidth causes data overloading. data overloading can be mitigated by implementing a potential solution to deal with the massive network traffic. Despite the fact that the capabilities of mobile networks are rapidly expanding, the aforementioned issue persists as we move from cellular to advanced wireless networks. To keep up with the ever-increasing number of internet users, offloading techniques have become increasingly popular in mobile computing. As the number of users and the volume of data they generate grows, the network’s capacity is taxed, and this is where offloading techniques come in to help. However, in many situations, authentication is absolutely necessary for the formation of a connection. As a result, the study presents a data offloading method for mobile computing that is both safe and cost-effective, as well as successful in managing network traffic. Individuals’ personal information might also constitute a severe privacy risk to users. Anticipating the legitimacy of the nodes and secret key negotiation helps avoid these dangers. Developing a mutual authentication and secret key exchange system for resource-constrained networks that is secure and private in all ways is a difficult challenge. Integrate key exchange authentication with an efficient data offloading approach to maintain data confidentiality and integrity in the whole network during transmission is the primary goal of the research. This research presents a Unique Tokenized Authentication Technique based data Offloading approach using Key Exchange (UTAT-DO-KE) model is proposed in this research for maintaining data integrity in heterogeneous networks during data communication. In this case, performance analysis can be done to lower the time it takes to authenticate and run the application. W

关键词： data integrity data offloading data overloading Key exchange Network traffic Unique token

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：