检索结果-内蒙古大学图书馆

Distributed Deep Learning with Data Parallelism for Diabetic Retinopathy Classification 1

7th International Conference on Smart Trends in computing and Communications, SmartCom 2023

作者： Patil, Sanskruti S. Patil, Mahesh S. Chickerur, Satyadhyan Giraddi, Shantala Shahapur, Seetharam N. Hadalageri, Anup School of Computer Science and Engineering KLE Technological University Karnataka Hubballi580031 India Center for High Performance Computing KLE Technological University Karnataka Hubballi580031 India

ISBN: (数字)9789819907694

ISBN: (纸本)9789819907687

The large-scale model training is typically slower and necessitates high-performance workstations and Distributed Deep Learning (DDL). The DDL models trained on a massive volume of data can outperform single accelerators being used, i.e., the performance of deep learning (DL) models can be enhanced by using distributed and parallel deep learning methods. But the DDL models need to be redesigned and evaluated for particular hardware resources and applications, requiring the literature on various DDL model evaluations for assistance. Authors in the proposed work design and evaluate DDL with data parallelism for classifying Diabetic retinopathy (DR) images. A performance comparative study on the DDL strategy implemented on DenseNet is presented where the model is trained on multi-GPUs with counts from 1 to 8. The results show that the model takes less time when trained on multi-GPUs compared to single GPU. The detailed evaluation analysis of DDL with data parallelism is presented in this paper. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Towards Developing an Ontology for Safety of Navigation Sensors in Autonomous Vehicles 15

Towards Developing an Ontology for Safety of Navigation Sens...

引用

15th International Conference on Knowledge engineering and Ontology Development, KEOD 2023 as part of the 15th International Joint Conference on Knowledge Discovery, Knowledge engineering and Knowledge Management, IC3K 2023

作者： Alharbi, Mohammed Karimi, Hassan A. Geoinformatics Laboratory School of Computing and Information University of Pittsburgh PittsburghPA15260 United States College of Computer Science and Engineering Taibah University Medina42353 Saudi Arabia

ISBN: (纸本)9789897586712

Understanding and handling uncertainties associated with navigation sensors in autonomous vehicles (AVs) is vital to enhancing their safety and reliability. Given the unpredictable nature of real-world driving environments, accurate interpretation and management of such uncertainties can significantly improve navigation decision-making in AVs. This paper proposes a novel semantic model (ontology) for navigation sensors and their interactions in AVs, focusing specifically on sensor uncertainties. At the heart of this new ontology is understanding the sources of sensor uncertainties within specific environments. The ultimate goal of the proposed ontology is to standardize knowledge of AV navigation systems for the purpose of alleviating safety concerns that stand in the way of widespread AV adoption. The proposed ontology was evaluated with scenarios to demonstrate its functionality. Copyright © 2023 by SCITEPRESS – science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0).

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

A Deep Reinforcement Learning Perspective on Adaptive Federated Dropout 20

A Deep Reinforcement Learning Perspective on Adaptive Federa...

引用

20th IEEE International Conference on Mobile Ad Hoc and Smart Systems, MASS 2023

作者： Sun, Tonghui Huang, Yan Xie, Zhenzhen Pang, Junjie Chen, Xingyun Cai, Zhipeng Qingdao University College of Computer Science and Technology Qingdao266000 China Kennesaw State University College of Computing and Software Engineering AtlantaGA United States Shandong University School of Computer Science and Technology Qingdao266237 China Georgia State University Department of Computer Science AtlantaGA United States

ISBN: (纸本)9798350324334

Federated learning is a distributed learning solution that achieves high-quality machine learning models while ensuring privacy and collaboration among various end devices. However, different kinds of end devices can lead to an unstable training process due to limited and dynamic communication and computation resources. Federated Dropout is an important technique for mitigating such resource bottlenecks by randomly removing a fixed percentage of activations of model components. However, most of the existing FL dropout methods cannot effectively utilize the distinguished characteristics of different clients or fail to automatically adapt to end device heterogeneity. In this paper, we thus propose Adaptive Federated Dropout with Reinforcement Learning (AFD-RL). AFD-RL employs reinforcement learning technology to adapt to the clients' unbalanced computation and communication resources and provide a personalized dropout strategy for each client. By automatically selecting suitable model components for each client in each communication round, AFD-RL achieves better accuracy with fewer communication rounds. Extensive experiments validate the effectiveness of the proposed approach in improving training efficiency and inference accuracy. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

An Efficient Graph Accelerator with Distributed On-Chip Memory Hierarchy 22nd

An Efficient Graph Accelerator with Distributed On-Chip Mem...

引用

22nd International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2022

作者： Zheng, Ran Jiang, Yingxin Wang, Yibo Su, Yongbo Zheng, Long Yao, Pengcheng Liao, Xiaofei Jin, Hai National Engineering Research Center for Big Data Technology and System/Services Computing Technology and System Lab/Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China Zhejiang Lab Hangzhou311121 China

ISBN: (纸本)9783031226762

Graph processing has evolved and expanded swiftly with artificial intelligence and big data technology. High-Bandwidth Memory (HBM), which delivers terabyte-level memory bandwidth, has opened up new development possibilities for FPGA-based graph accelerators. However, despite the tremendous expansion of underlying hardware capabilities, existing graph accelerators have not benefited too much. In this paper, we observe that the uniformed on-chip memory hierarchy is the key to the low scalability of existing graph accelerators. We present a novel graph accelerator with a distributed on-chip memory hierarchy called GraphS. The on-chip memory of GraphS is divided into numerous tiny blocks, each of which is assigned to only one Processing Element (PE). Different PEs are connected through a scalable network-on-chip (NoC). For realistic graphs with power-law properties, a degree-aware preprocessing method is designed to balance the workload among different PEs. Our results with various graph algorithms demonstrate that GraphS can outperform state-of-the-art ForeGraph by up to 21.84 ×. © 2023, Springer Nature Switzerland AG.

关键词： Network-on-chip

来源：评论

学校读者我要写书评

暂无评论

Comparison of Novel Machine Learning Algorithms for Predicting Chronic Renal Disease

Comparison of Novel Machine Learning Algorithms for Predicti...

引用

2024 International Conference on Computational Intelligence for Green and Sustainable Technologies, ICCIGST 2024

作者： Hima Bindu, G.B. Valluru, Dinesh Vachhani, Harshita Sunil Kumar, M. Ganesh, D. Neelima, P. The Apollo University School of Technology Department of of CSE Andhra Pradesh Chittoor India MLRITM Engineering College Department of IT Telangana Hyderabad India Pratibha College of Commerce and Computer Studies Chinchwad Dept.of Computer Science Maharshtra Pune India School of Computing Andhra Pradesh Tirupati India SPMVV School of Engineering & Technology Department of CSE Tiruapti India

ISBN: (纸本)9798350381887

Chronic Renal Disease (CRD) is a serious sickness that affects the entire world and is the primary cause of mortality for many other illnesses. Due to the lack of visible symptoms at the beginning of CRD, people usually neglect the illness. Early CRD diagnosis enables individuals to receive prompt therapy to slow the course of their illness. Because it quickly and accurately identifies potentials, model-based machine learning can help doctors accomplish this aim. In this paper, an ML strategy for CRD forecasting. The UCI machine learning repository provided the CRD dataset, which has a high number of uncertain data values. To address unknown values or unknown data, KNN estimation was employed, which picks multiple instances that contain the most comparable measures to manage the undetermined data value for every insufficient sample. Missing data is prevalent in real therapy situations, therefore individuals may skip particular measurements for a variety of reasons. The technique is generated using five machine learning algorithms (SVM, Logistic Regression, k-nearest neighbor, Random Forest, and Neural Network). Random forest outscored the other models of machine learning regarding accuracy. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A New Cloud Resource Manufacturing System Based on the Internet of Things Using Optimized Deep Neural Network 1

A New Cloud Resource Manufacturing System Based on the Inter...

引用

1st International Conference on Cognitive computing and engineering Education, ICCCEE 2023

作者： Supriya Menon, M. Spurthi, K. Elanangai, V. Nagaraju, D. Ashokkumar, C. Sundaram, Ajith Koneru Lakshmaiah Education Foundation Computer Science and Engineering Vaddeswaram AP Guntur522501 India Swarna Bharathi Institute of Science and Technology Computer Science and Engineering Khammam507002 India St. Peter's Institute of Higher Education and Research AVADI Department of Electrical and Electronics Engineering Chennai54 India Sri Venkatesa Perumal College of Engineering and Technology Department of CSE Andhra Pradesh Puttur India SRM Institute of Science and Technology School of Computing Department of Computing Technologies Kattankulathur Chennai603203 India Amrita Vishwa Vidyapeetham Amrita School of Business Kerala Kochi India

ISBN: (纸本)9798350332803

Cloud computing (CC) and the Internet of Things (IoT) are important in the modern era for providing new techniques of connection through intelligence perception, M2M, sharing efficient resources, and on-demand use. In a cloud environment, the manufacturing frame has numerous action tasks to perform or manufacture. Various Cloud Manufacturing system (CMfgs) resources influence some difficulties and attacks in manufacturing resources (work units, resource pools, workstations, and so on) and system modeling. To address these issues and meet the deadline, CMfgs proposed a novel Hybrid Grey Wolf-Ant Lion Regression Frame (HGW-ALRF) deep learning technique. In this case, the fitness function of the Ant Lion and Grey Wolf in the recurrent frame is used to effectively schedule cloud resources. During testing, the presented attacks (malware injection, DDoS, bottleneck attacks) can be monitored by the designed IoT application. Finally, the proposed model's obtained results are described, including computation time, accuracy, energy consumption, delay, and resource utilization. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Multitask Optimization in Real-World Applications: A Survey

引用

Journal of Artificial Intelligence and Technology 2023年第1期3卷 32-38页

作者： Yue Wu Hangqi Ding Benhua Xiang Jinlong Sheng Wenping Ma Kai Qin Qiguang Miao Maoguo Gong School of Computer Science and Technology Xidian UniversityXi’anChina School of Artificial Intelligence Xidian UniversityXi’anChina Department of Computing Technologies Swinburne University of TechnologyHawthornAustralia School of Electronic Engineering Xidian UniversityXi’anChina

Because of its strong ability to solve problems,evolutionary multitask optimization(EMTO)algorithms have been widely studied *** algorithms have the advantage of fast searching for the optimal solution,but it is easy to fall into local optimum and difficult to *** evolutionary multitask algorithms with evolutionary optimization algorithms can be an effective method for solving these *** the implicit parallelism of tasks themselves and the knowledge transfer between tasks,more promising individual algorithms can be generated in the evolution process,which can jump out of the local *** to better combine the two has also been studied more and *** paper explores the existing evolutionary multitasking theory and improvement scheme in ***,it summarizes the application of EMTO in different ***,according to the existing research,the future research trends and potential exploration directions are revealed.

关键词： evolutionary multitasking evolutionary algorithm optimization

来源：评论

学校读者我要写书评

暂无评论

Clustering-Based Analysis on 5G Networks: A Comparison 4

Clustering-Based Analysis on 5G Networks: A Comparison

引用

4th International Conference on Electronics and Sustainable Communication Systems, ICESC 2023

作者： Annam, Sangeetha Mithillesh Kumar, P. Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Amrita School of Computing Bengaluru Amrita Vishwa Vidyapeetham Dept. of Computer Science and Engg India

ISBN: (纸本)9798350300093

To satisfy the growing capacity demand and provide consumers with a higher level of service, it is anticipated that future-generation wireless networks will run entirely automatically. Utilizing spatiotemporal data on users and network health, mobile wireless networks may infuse the system with end-to-end visibility and intelligence. To handle the shifting conditions and the delivery of complex requirements for complicated services in a digital world with a highly engaged machine ecology, the future cellular system must be a neural carrier. By enabling the network to train and act, artificial intelligence enabled by '5G technology' can improve the network's interactive and decision-making capabilities. This study offers a thorough analysis of current clustering models in 5G networks. © 2023 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Multi-stage generative adversarial network model for segmenting retinal vascular structures in eye disease prediction

引用

Journal of Medical engineering and Technology 2025年 1-26页

作者： Bhanuse, Roshan S. Yenurkar, Ganesh Singh, Kavita R. Mal, Sandip Mane, Sulakshana B. Kachhwah, Rahul Rajbhar, Neeraj Take, Saksham Thakre, Tejas Department of Computer Technology Yeshwantrao Chavan College of Engineering Maharashtra Nagpur India School of Computing Science and Engineering VIT Bhopal University Bhopal India Department of Computer Engineering Bharati Vidyapeeth College of Engineering Maharashtra Navi Mumbai Mumbai India

Retinal vessel segmentation is essential for precise ophthalmological diagnoses, particularly in the prediction of retinal degenerative diseases. However, existing methods usually lack robustness and accuracy, especially in segmentation of thin or overlapping vessels. To face these challenges, this study introduces an enhanced retina-RV-Gain segmentation model, which employs an architecture of various stages to refine the results of segmentation iteratively. The model integrates attention mechanisms to better capture complex vessel structures and employs an adaptive loss function to manage class imbalance. In addition, a specially designed discriminator enhances the model’s ability to distinguish fine details from background noise vessels. The proposed RV-Gan is trained in comprehensive data sets that comprise retinal images, segmentation masks and noted labels, including Stare-DB, Chase-DB1 and Drive, using the Python platform. Experimental results demonstrate a segmentation accuracy of up to 99% in normal, abnormal and base vessels. These findings highlight the potential of the model to significantly improve diagnostic accuracy and support early prediction of disease in clinical ophthalmology. Overall, the enhanced RV-Gan architecture offers a robust solution to the limitations of current approaches, providing segmentation of high fidelity retinal vessels and advancing the predictive analysis of retinal degenerative conditions. © 2025 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Deep Convolutional Neural Networks for Automated Screening and Diagnosis of Ophthalmic Diseases

Deep Convolutional Neural Networks for Automated Screening a...

引用

Advancement in Technology (ICONAT), International Conference for

作者： Tina Babu Pramod Kumar Naik Rekha R Nair P G Janhavi Department of Computer Science and Engineering Alliance School of Advanced Computing Alliance University Bengaluru India Department of Computer Science Engineering Dayananda Sagar University Bengaluru India

ISBN: (数字)9798350354171

ISBN: (纸本)9798350354188

This project focuses on developing a deep learning algorithm to classify eye diseases from images. The dataset comprises images depicting various eye conditions, including glaucoma, cataracts, normal eyes, and diabetic retinopathy. We adopt a comprehensive approach by utilizing pre-trained architectures such as ResNet-18, GoogleNet, AlexNet, VGG19, and ResNet-50 as the backbone of our models, fine-tuning each on our dataset. During the training process, we employed data augmentation techniques to enhance the dataset, improving the models’ generalization ability across diverse eye conditions. The evaluation of the models involves a combination of cross-entropy loss and accuracy metrics to assess performance. Notably, the individually trained models exhibited varying accuracies, with ResNet-18 achieving the highest accuracy of $92.7 \%$ on the validation set. Additionally, we conducted an analysis of model performance using metrics such as precision, recall, and F1-score, further enhancing our understanding of the models’ capabilities. The visualization of results is facilitated through the presentation of confusion matrices, providing insights into the models’ classification behavior across different eye diseases. This study opens up a promising new path for improving ophthalmology screening and diagnosis by demonstrating the prospective benefits of deep learning in automating the classification of eye diseases.

关键词： Deep learning Measurement Glaucoma Diabetic retinopathy Accuracy Reviews Retina Feature extraction Convolutional neural networks Medical diagnostic imaging

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：