检索结果-内蒙古大学图书馆

20th IEEE India Council International Conference, INDICON 2023

作者： Bisht, Akhilesh Gupta, Deepa Parida, Shantipriya Amrita Vishwa Vidhyapeetham Amrita School of Computing Department of Computer Science and Engineering Bengaluru India Silo Ai Helsinki Finland

ISBN: (纸本)9798350305593

In the area of machine translation, the objective is to develop a model which can translate a piece of text from one language to another language. Large sized neural networks requires a large amount of training data to achieve a good performance level. The objective of this study is to build an explicit knowledge from the dependency parser, which can be integrated into transformer architecture. Adding an explicit knowledge to a neural machine translation network can enhance its learning process. The baseline transformer model and the proposed guided transformer model were tested on the same test data set of PMIndia Man Ki Baat session. With an addition of explicit knowledge, the proposed guided transformer model surpass the performance of baseline transformer model by +1.0 BLEU (approx.), +1.0 SacreBLEU (approx.), +1.48 Chrf, +1.44 Chrf+ and +1.39 Chrf++. © 2023 IEEE.

关键词： Blue-score Decoder Dependency Graph Dependency Parser Encoder English-Hindi Translation Explicit Knowledge Machine translation Multi head attention Positional encoding Self-attention Transformers

来源：评论

学校读者我要写书评

暂无评论

ai Approaches to Investigate EEG Signal Classification for Cognitive Performance Assessment 6

AI Approaches to Investigate EEG Signal Classification for C...

引用

6th International Conference on Computational Intelligence and Networks, CINE 2024

作者： Chatterjee, Rajdeep Chatterjee, Siddhartha Samanta, Saikat Biswas, Suman KIIT Deemed to be University School of Computer Engineering Bhubaneswar751024 India Amygdala AI Bhubaneswar751024 India College of Engineering and Management Department of Computer Science and Engineering Kolaghat721171 India Manipal University Jaipur Department of IoT & IS Rajasthan Jaipur303007 India

ISBN: (纸本)9798331516796

Scientists utilize Electroencephalogram (EEG) brain waves to evaluate different cognitive processes. It can involve the assessment of cognitive activities or the analysis of motorimagery motions. The EEG is a non-invasive and portable technique that offers exceptional temporal resolution. We have utilized two distinct EEG benchmark datasets to evaluate the efficacy of our suggested methods for extracting features and classifying data. We have utilized a range of well-known machine learning classifiers, such as long short-term memory (LSTM), one-dimensional convolutional neural networks (1D-CNN), and deep neural networks (DNN), to verify the model's accuracy both prior to and following feature selection. Our analysis reveals that the suggested process of extracting, selecting, and classifying features surpasses the current leading model in terms of performance. The suggested approaches achieve an accuracy of 93.36% and 98.65% for the EEGMAT and EEGMMIDB datasets, respectively. © 2024 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Revisiting Kernel Attention with Correlated Gaussian Process Representation 40

Revisiting Kernel Attention with Correlated Gaussian Process...

引用

40th Conference on Uncertainty in Artificial Intelligence, Uai 2024

作者： Bui, Long Minh Huu, Tho Tran Dinh, Duy Nguyen, Tan Minh Hoang, Trong Nghia FPT Software AI Center Department of Mathematics National University of Singapore Singapore School of Electrical Engineering and Computer Science Washington State University United States

Transformers have increasingly become the de facto method to model sequential data with state-of-the-art performance. Due to its widespread use, being able to estimate and calibrate its modeling uncertainty is important to understand and design robust transformer models. To achieve this, previous works have used Gaussian processes (GPs) to perform uncertainty calibration for the attention units of transformers and attained notable successes. However, such approaches have to confine the transformers to the space of symmetric attention to ensure the necessary symmetric requirement of their GP’s kernel specification, which reduces the representation capacity of the model. To mitigate this restriction, we propose the Correlated Gaussian Process Transformer (CGPT), a new class of transformers whose self-attention units are modeled as cross-covariance between two correlated GPs (CGPs). This allows asymmetries in attention and can enhance the representation capacity of GP-based transformers. We also derive a sparse approximation for CGP to make it scale better. Our empirical studies show that both CGP-based and sparse CGP-based transformers achieve better performance than state-of-the-art GP-based transformers on a variety of benchmark tasks. © 2024 Proceedings of Machine Learning Research. All rights reserved.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Prediction Model for Heart Disease Data Based on Feature engineering

Machine Learning Prediction Model for Heart Disease Data Bas...

引用

International Conference on Computing and Networking Technology (ICCNT)

作者： Md. Aamir Sohail Kunwar Babar Ali Satvik Vats Akhilesh Pandey Aditya Verma Department of CSE (AI) Meerut Institute of Engineering and Technology Meerut India Computer Science and Engineering Graphic Era Hill University Dehradun India Department of CSE (AI and ML) Meerut Institute of Engineering and Technology Meerut India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Growing health risks has brought up remarkable impact in the mind of people to find the most important factors responsible for diseases like diabetes, heart disease and so on. In this study, authors have gone through a set of factors associated with daily habits of people around us and used those factors to classify the risk of heart diseases by introducing feature engineering concepts in the form of grouping of categorical and numerical features. the paper focuses on the improvement of results in the form of accuracy through the introduction of a new model framework whose mathematical expression and detailed explanation is described in Methodology section. Apart from classification, it has also been studied about the inter-dependence of available attributes in the data. The grouping of features have been done through a well-defined algorithm described in this study.

关键词： Heart Accuracy Computational modeling Training data Machine learning Predictive models Encoding Data models Mathematical models Diseases

来源：评论

学校读者我要写书评

暂无评论

Semantic-Aware Pseudo-Labeling for Unsupervised Meta-Learning

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5475-5488页

作者： Ouyang, Tianran Dong, Xingping Ye, Mang Du, Bo Shao, Ling Shen, Jianbing Wuhan University School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan430072 China University of Chinese Academy of Sciences UCAS-Terminus AI Lab Beijing101408 China University of Macau State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science 314100 China

In unsupervised meta-learning, the clustering-based pseudo-labeling approach is an attractive framework, since it is model-agnostic, allowing it to synergize with supervised algorithms to learn from unlabeled data. However, the pseudo-labels suffer from clustering noise and semantic chaos problems, further impacting the effectiveness of meta-learning. In this paper, we analyze and optimize the pseudo-labeling process, including encoding and clustering, aiming to generate semantic-like pseudo-labels to narrow the gap between unsupervised and supervised meta-learning. First, during the encoding, we observe that the embedding space of existing methods lacks clustering-friendly properties, which is the primary reason for clustering noise. To address this issue, we minimize the inter-to-intra-class similarity ratio to generate clustering-friendly embedding features and validate our approach through comprehensive experiments. Then, during the clustering, we find that the semantic quality of pseudo-labels is not adequately controlled, resulting in semantic chaos of pseudo-labels. We propose a semantic-stability index to measure the semantic quality of pseudo-labels quantitatively. Based on this index, we propose the Semantic-aware Pseudo-label Reassignment mechanism to generate semantic-like pseudo-labels for all samples. Our approach is model-agnostic and can easily be integrated into existing supervised methods. To demonstrate its generalization ability, we integrate it into two representative algorithms: MAml and EP. The results on three main few-shot benchmarks clearly show that the proposed method achieves significant improvement compared to state-of-the-art models. Notably, our approach also outperforms the corresponding supervised method in three tasks. © 1979-2012 IEEE.

关键词： Latent semantic analysis

来源：评论

学校读者我要写书评

暂无评论

Towards Practical Edge Inference Attacks Against Graph Neural Networks 48

Towards Practical Edge Inference Attacks Against Graph Neura...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Li, Kailai Sun, Jiawei Chen, Ruoxin Ding, Wei Yu, Kexue Li, Jie Wu, Chentao Shanghai Jiao Tong University MoE Key Lab of Artificial Intelligence AI Institute Department of Computer Science and Engineering China

ISBN: (纸本)9781728163277

Graph Neural Networks (GNNs) have demonstrated superior performance in numerous real-world applications. Despite their success, recent studies have shown that GNNs are vulnerable under edge inference attacks aimed to infer the connectivity of a given pair of nodes. However, existing methods primarily focus on the scenario when properties of target nodes are revealed. In this paper, we propose an edge inference attack in a more realistic and practical setting. In our threat model, the adversary cannot obtain properties of target nodes but can inject a single probing node and query the target GNN for its prediction. By connecting the probing and target nodes, the adversary can infer the connectivity of the target node pair based on the prediction of the probing node. Extensive experiments show that our attack performs comparably to ones that require properties of target nodes. And when given such auxiliary knowledge, our attack outperforms state-of-the-art methods. © 2023 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimal Synergic Deep Learning for COVID-19 Classification Using Chest X-Ray Images

引用

computers, Materials & Continua 2023年第6期75卷 5255-5270页

作者： JoséEscorcia-Gutierrez Margarita Gamarra Roosvel Soto-Diaz Safa Alsafari Ayman Yafoz Romany F.Mansour Departament of Computational Science and Electronic Universidad de la CostaCUCBarranquilla080002Colombia Biomedical Engineering Program Universidad Simón BolívarBarranquilla080002Colombia Department of Computer Science and AI Faculty of Computer Science and EngineeringUniversity of JeddahJeddahSaudi Arabia Department of Information Systems Faculty of Computing and Information TechnologyKing Abdulaziz UniversityJeddahSaudi Arabia Department of Mathematics Faculty of ScienceNew Valley UniversityEl-Kharga72511Egypt

A chest radiology scan can significantly aid the early diagnosis and management of COVID-19 since the virus attacks the *** X-ray(CXR)gained much interest after the COVID-19 outbreak thanks to its rapid imaging time,widespread availability,low cost,and *** radiological investigations,computer-aided diagnostic tools are implemented to reduce intra-and inter-observer *** lately industrialized Artificial Intelligence(ai)algorithms and radiological techniques to diagnose and classify disease is *** current study develops an automatic identification and classification model for CXR pictures using Gaussian Fil-tering based Optimized Synergic Deep Learning using Remora Optimization Algorithm(GF-OSDL-ROA).This method is inclusive of preprocessing and classification based on *** data is preprocessed using Gaussian filtering(GF)to remove any extraneous noise from the image’s ***,the OSDL model is applied to classify the CXRs under different severity levels based on CXR *** learning rate of OSDL is optimized with the help of ROA for COVID-19 diagnosis showing the novelty of the *** model,applied in this study,was validated using the COVID-19 *** experiments were conducted upon the proposed OSDL model,which achieved a classification accuracy of 99.83%,while the current Convolutional Neural Network achieved less classification accuracy,i.e.,98.14%.

关键词： Artificial intelligence chest X-ray COVID-19 optimized synergic deep learning preprocessing public health

来源：评论

学校读者我要写书评

暂无评论

Cost Optimization Approaches for Task and Workflow Scheduling in the Cloud Environment 15th

Cost Optimization Approaches for Task and Workflow Schedulin...

引用

15th International Conference on Soft Computing and Pattern Recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2023

作者： Mahendar, A. Senthamarai Kannan, S. Anantha Sivaprakasam, S. Santhosh Kumar, P.N. Pradeep Ghantasala, G.S. Kallam, Suresh CMR Technical Campus Telangana Hyderabad India Department of AI and Data Science SRM TRP Engineering College Irungalur Tamilnadu Thiruchirapalli India Department of Computer Science and Engineering Rajalakshmi Engineering College Thandalam Tamilnadu Chennai India CMR Technical Campus Hyderabad Telangana Hyderabad India Department of Computer Science and Engineering Alliance University Bengaluru India School of Computer Science and Engineering Jain Deemed to Be University Bengaluru India

ISBN: (纸本)9783031810855

The use of information technology by companies and individuals in cloud computing revolutionizes. Cloud computing offers virtualized tools over the Internet for dynamic services. This ensures that end-users with resources-virtualization maintained and controlled can create, deploy and manage applications "on the cloud." The preparation of tasks in cloud computing systems is one of the most challenging issues in fulfilling user-defined service quality while minimizing the cost of executing tasks. Several cost optimization approaches have been proposed to improve the economic aspect of task scheduling in cloud computing. A full analysis of approaches to promoting cost optimization in connection with task planning in cloud computing has not yet been published in the literature. This paper is intended to examine the problems of cost optimization in task scheduling through a detailed analysis of current task scheduling methods in cloud computing and to identify aspects and parameters of cost optimization. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Cost benefit analysis

来源：评论

学校读者我要写书评

暂无评论

Crowd Coordination System

Crowd Coordination System

引用

Disruptive Technologies (ICDT), International Conference on

作者： Sandeep Kumar Anushka Singhal Ishan Sangal Medhavi Bhardwaj Department of Computer Science & Engineering (AI) ABESIT Ghaziabad

Effective crowd management is Privacy-Ensuring public safety and optimizing the efficiency of diverse events and public spaces. In real time of The system can estimate crowd density and monitor changes over time by analyzing video feeds or images adequate for handling large crowds. The suggested system utilizes Y olov7 algorithms and computer vision techniques to accurately count individuals in real-time. Through the analysis of video feeds or images, the system can estimate crowd density and monitor changes over time. When predefined thresholds for crowd size or density are met, the system automatically generates alert emails using the Gmail API, sending them to pertinent stakeholders such as event organizers, security personnel, or public authorities. These alerts facilitate prompt responses to potential overcrowding situations, allowing for proactive crowd management and improved safety measures. This summary underscores the potential of ai -driven crowd management systems to offer real-time monitoring and alerting capabilities, contributing to effective crowd control and ensuring the welfare of individuals in various public spaces or events. The proactive approach aims to boost safety, streamline crowd flow, and mitigate risks associated with overcrowding.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Analysis on Virtual Mouse Control using Human Eye

An Analysis on Virtual Mouse Control using Human Eye

引用

Image Processing and Capsule Networks (ICIPCN), International Conference on

作者： P. M. Naidu N. Muthukumaran S. Chandralekha K. Tejaswini Reddy K.Sri Vaishnavi Department of Computer Science and Engineering (AI&ML) Vignan Institute of Technology and Science Department of ECE Centre for Computational Imaging and Machine Vision Sri Eshwar College of Engineering Coimbatore India

ISBN: (数字)9798350367171

ISBN: (纸本)9798350367188

Technological advances have enabled hands-free computer interaction, especially for people with limited hand function, such as amputees and paraplegics. However, current solutions face limitations in terms of input methods, accuracy, functionality, and accessibility. The proposed model aims to solve these problems by developing a reliable, user-friendly, and accurate system by utilizing facial expressions, especially eye and mouth movements. The proposed system aims to improve reliability and expressiveness by combining two modalities. Using advanced image processing technology, errors caused by lighting changes are minimized to provide a smooth user experience. The virtual mouse interface provides comprehensive features such as scrolling, clicking, and typing, improving the experience of users with different needs. By providing a reliable, versatile, and convenient hands-free interaction system, this study aims to give users greater independence, improve communication, and increase their participation in the digital world.

关键词： Accuracy Image processing computer network reliability Computational modeling Mouth Lighting User experience

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：