检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Wang, Haolin Liu, Xuefeng Niu, Jianwei Guo, Wenkai Tang, Shaojie State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Center for AI Business Innovation Department of Management Science and Systems University at Buffalo BuffaloNY United States Zhongguancun Laboratory Beijing China

Federated learning is a distributed machine learning paradigm designed to protect user data privacy, which has been successfully implemented across various scenarios. In traditional federated learning, the entire parameter set of local models is updated and averaged in each training round. Although this full network update method maximizes knowledge acquisition and sharing for each model layer, it prevents the layers of the global model from cooperating effectively to complete the tasks of each client, a challenge we refer to as layer mismatch. This mismatch problem recurs after every parameter averaging, consequently slowing down model convergence and degrading overall performance. To address the layer mismatch issue, we introduce the FedPart method, which restricts model updates to either a single layer or a few layers during each communication round. Furthermore, to maintain the efficiency of knowledge acquisition and sharing, we develop several strategies to select trainable layers in each round, including sequential updating and multi-round cycle training. Our theoretical analysis and experimental results show that the FedPart method significantly outperforms traditional full-network update strategies, achieving faster convergence, greater accuracy, and reduced communication and computational overhead. © 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

Dynamic Resource Allocation-Enabled Distributed Learning as a Service for Vehicular Networks

Dynamic Resource Allocation-Enabled Distributed Learning as ...

引用

Data science and Information System (ICDSIS), IEEE International Conference on

作者： Thirusubramanian Ganesan Ramy Riad Al-Fatlawy Suma Srinath Srinivas Aluvala R. Lakshmana Kumar Cognizant Technology Solutions Texas USA Department of Computers Techniques Engineering College of Technical Engineering The Islamic university Najaf Iraq Department of AIML Nitte Meenakshi Institute of Technology Bengaluru India School of Computer Science and Engineering SR University Warangal Telangana India Department of Computer Science and Engineering (AI & ML) Tagore Institute of Engineering & Technology Salem Tamil Nadu India

ISBN: (数字)9798350381665

ISBN: (纸本)9798350381672

The growth and development of 6 G networks bring a crucial task onto the table which is introducing ai to manage network functions and cope with more users in the future. Distributed Learning (DL), as service to technologies, acts as one of the overriding forces motivating the development of 6 G communication system. The 6 G will help in various applications, which are a large set of services; then you will be able to manage the network in a smart and dynamic way. While relatively, many deep learning (DL) methods is deployed within the resource constrained vehicular domain, this setting nonetheless poses significant challenges. The reciprocating development of distributed computing and communication resources, like the edge-cloud continuum and integrated terrestrial-non-terrestrial networks (T/NTN), offers a solution for this. To use these resources and multiple DL methods in an integrated way, the NS becomes the best option. This paper covers the methods of DL that apply best in vehicular environments and examines NS roles alongside these methods, especially in the context of resource management which is dynamic. This research is designed to build a architecture for DL-as-a-Service (DLaaS) letting it to be hosted on a distributed network platform and allowing the implementation of DL algorithms at a proactive basis. This method, in addition to the dynamic resource allocation strategies, permits the efficient control of the different services which have different requirements. Effectiveness of the model is exhibited in a full case step by step in a vehicular T/NTN. The run-down of the DLaaS approach is perceived from its strong points, namely flexibility, performance upgrades, improved network intelligence, and they lead eventually to heightened customer satisfaction from the given cross vehicle or non cross vehicle highway traffic scenarios.

关键词： computer aided instruction Distance learning Urban areas Signal processing algorithms Signal processing Dynamic scheduling Throughput

来源：评论

学校读者我要写书评

暂无评论

Support Vector Regression Based Traffic Prediction Machine Learning Model*

Support Vector Regression Based Traffic Prediction Machine L...

引用

International Conference on Computation System and Information Technology for Sustainable Solutions (CSITSS)

作者： Ansh Srivastava Mrigaannkaa Singh Somesh Nandi Dept. of Computer Science and Engineering RV College of Engineering Bengaluru India Dept. of Electronics and Communication RV College of Engineering Bengaluru India Dept. of AI/ML RV College of Engineering Bengaluru India

ISBN: (数字)9798331505462

ISBN: (纸本)9798331505479

To combat the rising issue of urban traffic congestion, optimization of traffic light timings at intersections is crucial. Traditional methods show stagnancy in managing traffic flow dynamically. This study introduces Support Vector Regression (SVR) as an unconventional solution for traffic signal optimization. The model predicts the traffic conditions and alters the green light to turn on and off accordingly. Utilization of a comprehensive dataset consisting of vehicle distributions and critical flow ratios have been implemented to training a machine learning model. The statistics and performance is checked by usage of Mean Squared Error, R Squared, Mean Absolute Error demonstrating that SVR can effectively enhance traffic signal control. This approach pledges significant improvements in reduction in congestion and streamlined traffic flow. Our findings majorly highlight the capability of the SVR Machine Learning model in improving urban traffic management systems and call for further enhancement and collaboration with experts to address dynamic traffic issues.

关键词： Support vector machines Training Computational modeling Urban areas Machine learning Predictive models Vectors Vehicle dynamics Optimization Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Signbridge-Audio to Sign Language Translator

Signbridge-Audio to Sign Language Translator

引用

IEEE Students' Conference on Electrical, Electronics and computer science (SCEECS)

作者： Kammadanam Shirisha E. Mani Deeksith M. S. S Sri Madhava S. Surendhar R. Karthikeyan Department of Computer Science and Engineering Vardhaman College of Engineering Hyderabad India Department of CSE(AI&M) Vardhaman College of Engineering Hyderabad India

ISBN: (数字)9798331529833

ISBN: (纸本)9798331529840

The lack of communication options for Deaf and hearing people, some may say creates a significant social disadvantage in accessing the often-bare essential services. In contrast to acoustically communicated sound patterns, sign language communicates ideas freely through manual and body language. It can be used by those who have trouble speaking, by those who can hear but cannot talk, and by regular people to interact with those with hearing impairments. By automatically converting spoken words into hand gestures, this project creates a web-based interface that allows hearing-impaired individuals to communicate with normal people in real time through sign language interpretation. Two major phases of the system Speech-to-text technology interprets oral input into textual output. Next, the text is run through Natural Language Processing (NLP) algorithms leveraging the Natural language toolkit (NLTK) to be syntactically parsed in the context of sign natural rules. The last phase translates the parsed text into sign language gestures which involve hand shapes, orientation, and body movements to convey the visual meaning of a message. This system could be used to greatly reduce the communication barriers experienced by people with hearing loss and deafness using Machine Learning (ml) to continuously improve accuracy, enhance quality of life, and provide a more inclusive society between the deaf community in our daily interactions within an ever-growing world.

关键词： Hands Sign language Visualization Translation Semantics Auditory system Syntactics Rendering (computer graphics) Natural language processing Speech to text

来源：评论

学校读者我要写书评

暂无评论

Histopathology Image Analysis Utilizes Homology-Based Feature Extraction for Accurate Diagnosis of Various Lung Diseases 2

Histopathology Image Analysis Utilizes Homology-Based Featur...

引用

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

作者： Alabdeli, Haider Chandra, N Subhash Sowmya, M. Bhavani, R. Kumar, Bura Vijay The Islamic University Najaf Iraq Cvr College of Engineering Telangana India Nitte Meenakshi Institute of Technology Department of Ai and Ds Bengaluru India Sathyabama Institute of Science and Technology Chennai India Sr University School of Computer Science and Artificial Intelligence Warangal India

ISBN: (纸本)9798350317558

Histopathology image analysis is pivotal for diagnosing lung diseases, providing critical insights into tissue abnormalities for accurate medical assessments. Despite its significance, challenges arise from interpretational subjectivity and handling diverse tissue types, limiting its widespread application. The lung and colon histopathology dataset, comprising 25,000 color images, serves as a valuable resource in comprehensive medical research, offering diverse samples for detailed analysis and diagnosis. The proposed method introduces innovation by leveraging homology-based image processing for feature extraction, integrating the Unsharp Masking method in preprocessing. Classification through a Convolutional Neural Network (CNN) ensures a thorough and accurate analysis of histopathological images in lung disease diagnosis. Comparative analysis with existing approaches like 2D wavelet features and CRCNN-net underscores the undeniable superiority of the proposed method. The results demonstrate exceptional performance metrics, with an accuracy of 98.67%, precision at 93.70%, an F1-score of 80.22%, recall reaching 89.76%, and specificity standing at 90.82%. These outcomes affirm the proposed method's efficacy in advancing the field of histopathological lung disease diagnosis. © 2024 IEEE.

关键词： Disease diagnosis Histopathology image Homology based image Lung diseases Unsharp masking

来源：评论

学校读者我要写书评

暂无评论

Addressing Concerns: The Potential Impact of Big Informatics on Farmers 15

Addressing Concerns: The Potential Impact of Big Informatics...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Sharma, Manish Akshita Malik, Suraj Yuvaraj, K. Yawanikha, T. Nawadkar, Ashwini Ranjit Graphic Era Deemed to be University Department of Computer Science & Engineering Uttarakhand Dehradun India Invertis University Department of Agriculture Uttar Pradesh Bareilly243123 India Iimt University School of Engineering and Technology Meerut India Karpagam Academy of Higher Education Department of Computer Science Tamil Nadu Coimbatore India Karpagam Institute of Technology Department of Information Technology Tamil Nadu Coimbatore India Vishwakarma Institute of Technology Ai & Ds Pune India

ISBN: (纸本)9798350370249

This paper explores the possibilities and challenges of big data and comes to the conclusion that these techniques will enable pertinent research throughout the entire farming production process. Farther than scale, there are other qualities that characterize big data, including data amount, pace, diversity, and truth. We go over a group of analysis methods that become more and more important to our line of work as one deals with these problems. In the end, we come to the conclusion that farming and field economics are ideally situated to add to the Big Data study and dissemination goal. We think that our field could even make significant advances to pertinent policy, crop production, supply chain, customer demand, and environmental problems. The writers appreciate the constructive criticism provided by the unnamed evaluators and manager Trevor Verdier. The State Agribusiness and Woodland Trial Laboratory Special Investigation supported the *** prevalence of devices, geolocation equipment, and video surveillance has likely altered many agricultures production as well as the administration of the food industry as it moves from farmers to customers, according to a number of signs. © 2024 IEEE.

关键词： Analysis methods Big Data Data amount Data diversity Data pace Data truth Economic research priorities Farm economics Farm management Farmers Farming production process Precision agriculture Smart farming

来源：评论

学校读者我要写书评

暂无评论

Fast Vocabulary Transfer for Language Model Compression

arXiv

引用

arXiv 2024年

作者： Gee, Leonidas Rigutini, Leonardo Zugarini, Andrea Torroni, Paolo Expert.ai Italy Expert.ai University of Siena Italy Department of Computer Science and Engineering University of Bologna Italy

Real-world business applications require a trade-off between language model performance and size. We propose a new method for model compression that relies on vocabulary transfer. We evaluate the method on various vertical domains and downstream tasks. Our results indicate that vocabulary transfer can be effectively used in combination with other compression techniques, yielding a significant reduction in model size and inference time while marginally compromising on performance. Copyright © 2024, The Authors. All rights reserved.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Arabic Tweets-Based Sentiment Analysis Using Optuna-Tuned Bidirectional Gated Recurrent Unit Classifier 2

Arabic Tweets-Based Sentiment Analysis Using Optuna-Tuned Bi...

引用

2nd IEEE International Conference on Networks, Multimedia and Information Technology, NMITCON 2024

作者： Ashwini Kumari, P. Kumar, Komuravelly Sudheer Patil, Rajeshwari Adnan, Myasar Mundher Soujanya, Rajireddy Reva University Department of Eee Bengaluru India Sr University School of Computer Science & Artificial Intelligence Warangal India Nitte Meenakshi Institute of Technology Department of Ai & Ds Bengaluru India The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq Gokaraju Rangaraju Institute of Engineering and Technology Department of Cse Hyderabad India

ISBN: (纸本)9798350372892

Arabic sentiment analysis (ASA) is a dynamic field in Natural Language Processing (NLP) that applies Machine Learning (ml) techniques to understand the complex emotional tones that are present in Arabic text. By using ml techniques for sentiment analysis in Arabic, it can gain a deeper insight into human interactions within the verbal context. ASA is quite challenging due to its structural diversity and dialects. The proposed method uses a Bi-GRU classifier and optuna optimizer to enhance the ability to understand and analyze sequential data. The approach involves using a labeled Twitter dataset, pre-processing techniques, feature extraction, optuna optimization, and Bi-directional Gated Recurrent Unit (Bi-GRU) as a classifier. The results of this paper clarify the efficiency of DL models in accurately classifying sentiments in Arabic text, as illustrated by performance metrics such as 93.86® precision, 93.56® recall, 94.08® accuracy, and 93.88 F1 measure which is greater than the other existing methods like stacking logistic regression, Modified Switch Transformer (MST) and Bidirectional Long Short-Term Memory+ Dependency-based rules word embedding. © 2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

ml-Orchestrated Malware Detection Framework for Windows APIs in Critical Infrastructure Systems

ML-Orchestrated Malware Detection Framework for Windows APIs...

引用

computer, Communication and Control (IC4), International Conference on

作者： Shayalkumar Vaghasiya Rajesh Gupta Santoshinee Mohapatra Sudeep Tanwar Deepak Garg Department of Computer Science and Engineering Institute of Technology Nirma University Gujarat India School of Computer Science and AI SR University Warangal Telangana

The security of critical infrastructure systems is crucial as they provide basic amenities and services necessary for functioning of a community, also they represent important part of country’s economy. An attack on this critical infrastructure systems can have significant consequences. In this paper we discuss the This paper addresses the increasing security challenges in critical infrastructure systems due to cyber threats, particularly malware and propose a framework based on machine learning for robust malware classification. Our proposed framework includes a dataset of Windows API functions used by the software, preprocessing techniques, and the XGBoost model which achieves an accuracy of 86.98%. Using this model, we can detect and categorize five different types of malwares to provide security against malware attacks. The results show model’s ability to successfully classify different kinds of malware. Realtime monitoring is another feature of this framework which is important for actively protecting critical infrastructure systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An ai Based Chat Bot For Banking Applications using Intelligent Chat Bots

An AI Based Chat Bot For Banking Applications using Intellig...

引用

Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication engineering (RMKMATE), International Conference on

作者： Ajmeera Kira Dbk Kamesh D Paulraj Veda Vidhya Srilakshmi V Department of Computer Science and Engineering MLR Institute of Technology Hyderabad India Department of CSE Malla Reddy Engineering College for Women Hyderabad Department of Computer Science and Engineering R.M.K. Engineering College Tamil Nadu India Department of Computer Science and Engineering MLR Institute of Technology Hyderabad Department of CSE(AI & ML) B V Raju Institute of Technology

A chat and voice assistant that is powered by ai and is integrated into a banking application. This assistant translates data in such a way that the text and audio are designed to simulate human conversations. Since their inception, voice and chat assistants have seen substantial technological advancement. The advancements that have been made in machine learning and natural language processing have made it possible for them to now learn from their experiences and conduct conversations that are more human-like. Because of their more human nature, they are able to assist with a wide variety of tasks, such as calculating loan interests, confirming facts regarding transactions, and checking savings.

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：