检索结果-内蒙古大学图书馆

IEEE international conference on Multimedia and Expo (ICME)

作者： Wu, Chenrui Li, Zexi Wang, Fangxin Wu, Chao Chinese Univ Hong Kong Future Network Intelligence Inst Shenzhen Peoples R China Chinese Univ Hong Kong Sch Sci & Engn Shenzhen Peoples R China Guangdong Prov Key Lab Future Networks Intelligen Shenzhen Peoples R China Peng Cheng Lab Shenzhen Peoples R China Zhejiang Univ Coll Comp Sci & Technol Hangzhou Peoples R China Zhejiang Univ Sch Publ Affairs Hangzhou Peoples R China

ISBN: (纸本)9781665468916

Federated learning (FL) is a distributed framework for collaborative training with privacy guarantees. In real-world scenarios, clients may have Non-IID data (local class imbalance) with poor annotation quality (label noise). The coexistence of label noise and class imbalance in FL's small local datasets renders conventional FL methods and noisy-label learning methods both ineffective. To address the challenges, we propose FEDCNI without using an additional clean proxy dataset. It includes a noise-resilient local solver and a robust global aggregator. For the local solver, we design a more robust prototypical noise detector to distinguish noisy samples. Further to reduce the negative impact brought by the noisy samples, we devise a curriculum pseudo labeling method and a denoise Mixup training strategy. For the global aggregator, we propose a switching re-weighted aggregation method tailored to different learning periods. Extensive experiments demonstrate our method can substantially outperform state-of-the-art solutions in mix-heterogeneous FL environments.

关键词： Federated learning Noisy labels Non-IID data Class imbalance

来源：评论

学校读者我要写书评

暂无评论

Towards a Collective Medical Imaging AI: Enabling Continual learning from Peers 7

Towards a Collective Medical Imaging AI: Enabling Continual ...

引用

7th international conference on Medical Imaging with Deep learning, MIDL 2024

作者： Zheng, Guangyao Braverman, Vladimir Leal, Jeffrey Rowe, Steven Leung, Doris Jacobs, Michael A. Parekh, Vishwa S. Department of Computer Science Rice University HoustonTX United States Department Of Diagnostic And Interventional Imaging McGovern Medical School UTHealth Houston HoustonTX United States The Russell H. Morgan Department of Radiology and Radiological Science The Johns Hopkins University School of Medicine BaltimoreMD21205 United States Center University of Maryland School of Medicine BaltimoreMD21201 United States Kennedy Krieger Institute 716 North Broadway Room 411 BaltimoreMD21205 United States Department of Neurology Johns Hopkins University School of Medicine BaltimoreMD United States Department of Radiology UNC School of Medicine Chapel HillNC United States

Federated learning is an exciting area within machine learning that allows cross-silo training of large-scale machine learning models on disparate or similar tasks in a privacy-preserving manner. However, conventional federated learning frameworks require a synchronous training schedule and are incapable of lifelong learning. To that end, we propose an asynchronous decentralized federated lifelong learning (ADFLL) method that allows agents in the system to asynchronously and continually learn from their own previous experiences and others’, thus overcoming the potential drawbacks of conventional federated learning. We evaluate the ADFLL framework in two experimental setups for deep reinforcement learning (DRL) based landmark localization across different imaging modalities, orientations, and sequences. The ADFLL was compared to central aggregation and conventional lifelong learning for upper-bound comparison and with a conventional DRL model for lower-bound comparison. Across all the experiments, ADFLL demonstrated excellent capability to collaboratively learn all tasks across all the agents compared to the baseline models in in-distribution and out-of-distribution test sets. In conclusion, we provide a flexible, efficient, and robust federated lifelong learning framework that can be deployed in real-world applications. © 2024 CC-BY 4.0, G. Zheng, V. Braverman, J.L., S.R., D.L., M.A. Jacobs & V.S. Parekh.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Defect Screening on Nuclear Power Plant Concrete Structures: A Two-staged Method Based on Contrastive Representation learning 8

Defect Screening on Nuclear Power Plant Concrete Structures:...

引用

8th IEEE international conference on Advanced Robotics and Mechatronics, ICARM 2023

作者： Huang, Wenlian Zhu, Guanming Huang, Qiang Chen, Zhuangzhuang Chen, Jie Li, Jianqiang College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen China

ISBN: (纸本)9798350300178

Intelligent defect detection methods are important for the surface of the containment of nuclear power plants and face many challenges in the field of computer vision. Due to the irregular shapes and large variation of defects, as well as the similarity of the features between some defects and the background. Most existing deep learning-based defect detection networks suffer from insufficient feature extraction capabilities, making it difficult to detect defects from the background. Inspired by discrete representation learning methods, we propose a two-stage defect detection model called ConVQVAE, which uses a semi-supervised method for representation learning to achieve binary classification for defect detection. In addition, contrastive learning is performed in two stages to enhance the model's diversity representation ability and inter-class disentanglement representation ability. Finally, we experimentally verify that the proposed method can obtain good classification results. © 2023 IEEE.

关键词： Defects

来源：评论

学校读者我要写书评

暂无评论

Selective Tensorized Multi-layer LSTM for Orbit Prediction 22

Selective Tensorized Multi-layer LSTM for Orbit Prediction

引用

31st ACM international conference on Information and Knowledge Management (CIKM)

作者： Shin, Youjin Park, Eun-Ju Woo, Simon S. Jung, Okchul Chung, Daewon Sejong Univ Dept Data Sci Seoul South Korea Sungkyunkwan Univ Dept Comp Sci & Engn Suwon South Korea Sungkyunkwan Univ Dept Appl Data Sci SKKU Inst Convergence Suwon South Korea KARI Natl Satellite Operat Ctr Daejeon South Korea

ISBN: (纸本)9781450392365

Although the collision of space objects not only incurs a high cost but also threatens human life, the risk of collision between satellites has increased, as the number of satellites has rapidly grown due to the significant interests in many space applications. However, it is not trivial to monitor the behavior of the satellite in real-time since the communication between the ground station and spacecraft is dynamic and sparse, and there is an increased latency due to the long distance. Accordingly, it is strongly required to predict the orbit of a satellite to prevent unexpected contingencies such as a collision. Therefore, the real-time monitoring and accurate orbit prediction are required. Furthermore, it is necessary to compress the prediction model, while achieving a high prediction performance in order to be deployable in the real systems. Although several machine learning and deep learning-based prediction approaches have been studied to address such issues, most of them have applied only basic machine learning models for orbit prediction without considering the size, running time, and complexity of the prediction model. In this research, we propose Selective Tensorized multi-layer LSTM (ST-LSTM) for orbit prediction, which not only improves the orbit prediction performance but also compresses the size of the model that can be applied in practical deployable scenarios. To evaluate our model, we use the real orbit dataset collected from the Korea Multi-Purpose Satellites (KOMPSAT-3 and KOMPSAT-3A) of the Korea Aerospace Research Institute (KARI) for 5 years. In addition, we compare our ST-LSTM to other machine learning-based regression models, LSTM, and basic tensorized LSTM models with regard to the prediction performance, model compression rate, and running time.

关键词： Tensor-train Decomposition Multi-layer LSTM Orbit Prediction Model Compression Parameter Reduction

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Training Optimizers on Diabetic Retinopathy Detection Using SP-Net CNN

Performance Analysis of Training Optimizers on Diabetic Reti...

引用

2024 international conference on Decision Aid sciences and Applications, DASA 2024

作者： Akhtar, Samia Aftab, Shabib Kousar, Shagufta Rehman, Ambreen Ahmad, Munir Saeed, Ali Q Virtual University of Pakistan Department of Computer Science Lahore Pakistan Applied Science Research Center Applied Science Private University Amman Jordan College of Dentistry University of Science and Technology of Fujairah Fujairah United Arab Emirates College of Informatics Korea University Seoul02841 Korea Republic of NCBA&E Lahore54000 Pakistan Computer Centre Northern Technical University Ninevah Iraq

ISBN: (纸本)9798350369106

Diabetic retinopathy is an ocular disease linked to long-term diabetes mellitus and it can lead to permanent loss of vision if not addressed promptly. Timely detection of this disease is crucial as it can significantly reduce the risk of severe complications. Our aim is to identify the most effective optimization algorithm for training our deep learning (CNN) model. Our goal is to achieve high accuracy and efficiency in grading diabetic retinopathy into 5 different classes. In this research, we conducted a comparative evaluation of three optimizers namely Adam, SGD and RMSProp. For classification, we developed a convolutional neural network called SP-Net. An image based dataset named APTOS 2019 is taken from Kaggle and various pre-processing techniques are applied on it such as cropping, denoising, histogram equalization, and resizing. We used data augmentation techniques to tackle the class imbalance issue. Then the dataset was divided into three subsets (training, validation, and testing) using a split of 70:10:20. Our model SP-Net incorporates several layers including convolutional, dropout, max pooling, and fully connected layers. The results demonstrate that SP-Net achieved a highest test accuracy of 97.76% when it was trained using the SGD optimizer. Moreover, this model trained with the best optimizer outperformed the existing state-of-the-art techniques on comparison. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Review of Optimal Energy Management Strategies Using machine learning Techniques for Hybrid Electric Vehicles

引用

international JOURNAL OF AUTOMOTIVE TECHNOLOGY 2021年第5期22卷 1437-1452页

作者： Song, Changhee Kim, Kiyoung Sung, Donghwan Kim, Kyunghyun Yang, Hyunjun Lee, Heeyun Cho, Gu Young Cha, Suk Won Seoul Natl Univ Dept Mech Engn Seoul 08826 South Korea Dankook Univ Dept Mech Engn Gyeonggi 16890 South Korea Seoul Natl Univ Inst Adv Machines & Design Seoul 08826 South Korea

A hybrid electric vehicle (HEV) is defined as a vehicle that has two or more power sources, the hybrid electric vehicle is a representative eco-friendly vehicle because it can operate efficiently with each power source and requires only a small sized electric power source. However, it is not possible to develop high efficiency HEVs without an effective energy management system (EMS), a well-designed EMS is vital in HEVs because they need to manage two power sources. Motivated by this, there are continuing efforts being made to research and establish suitable energy management strategies in order to develop high efficiency HEVs. In the past, many energy management strategies for HEVs were developed based on optimal control theory. Recently, various kinds of machine learning technologies have been applied to HEV EMS development based on breakthroughs in the fields of machine learning and artificial intelligence (AI). machine learning is a field of research that allows computers to perform arbitrary tasks guided by data rather than explicit programming. machine learning can be classified into supervised learning, reinforcement learning (semi-supervised learning), and unsupervised learning depending on how the training data is structured. In this study, we look at cases and studies in which machine learning techniques from each category were used to develop HEV energy management strategies.

关键词： Hybrid electric vehicle Energy management strategy machine learning Artificial intelligence Optimal control theory

来源：评论

学校读者我要写书评

暂无评论

Human-Like Networking Based on Big data and machine learning 11th

Human-Like Networking Based on Big Data and Machine Learning

引用

11th international conference on Signal and Information Processing, Network and Computers, ICSINC 2023

作者： Dai, Jinyou Yu, Shaohua Cheng, Chen Cheng, Xinzhou Lyu, Jianxin Wang, Xueshun China Information Communication Technology Group Wuhan China Pengcheng Laboratory Shenzhen China Research Institute China United Network Communications Corporation Beijing China Fiberhome Telecommunication Technologies Co. Ltd. Wuhan China

ISBN: (纸本)9789819721238

It can bring about innovation of network and new network architecture for big data technologies, artificial intelligence technologies, network awareness technologies, network self-restructuring technologies and other technologies to be applied into network area. Introducing human-like features into network can make the network make full use of advantages of human being and bring a new network architecture named human-like networking to come into being. This paper specifies the architecture, features and mechanism of human-like networking. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Crash prediction for freeway work zones in real time: A comparison between Convolutional Neural Network and Binary Logistic Regression model

引用

international JOURNAL OF TRANSPORTATION science AND TECHNOLOGY 2022年第3期11卷 484-495页

作者： Wang, Junhua Song, Hao Fu, Ting Behan, Molly Jie, Lei He, Yingxian Shangguan, Qiangqiang Minist Educ Engn Res Ctr Rd Traff Safety & Environm Shanghai Peoples R China Tongji Univ Coll Transportat Engn 4800 Caoan Highway Shanghai Peoples R China McGill Univ Dept Civil Engn & Appl Mech Quebec City PQ H3A 0C3 Canada Zhejiang Commun Investment Grp CO LTD Hangjinqu Branch Hangzhou Zhejiang Peoples R China 4800 Caoan Highway Shanghai Peoples R China

Safety of drivers in freeway work zones has been a problem. Real-time crash prediction helps prevent crashes before they happen. This paper looks at real-time crash prediction in freeway work zones by using machine learning approaches. Both the Convolutional Neural Network and the Binary Logistic Regression model are introduced. For training and testing the models, crash data and traffic data from several freeways in D7 zone, Los Angeles, California, were used. Crash data were collected from California Highway Patrol Incident System, and traffic data were obtained from the Caltrans Performance Measurement System. data processing and matching were conducted. Both the two models were trained and tested. Results show that the Convolutional Neural Network performed slightly better over the Binary Logistic Regression model in predicting crashes with a global accuracy of 79.50%. Despite this, the main merit of the Binary Logistic Regression model is that it is able estimate the impact of affecting variables on the probability of crashes and can help identify the factors related to risks in work zones. machine learning approaches applied in this study perform well in crash prediction. In general, machine learning techniques and reliable real-time crash prediction applications can be promising in helping drivers and transportation engineers make timely responses to potential crashes on freeways.& COPY;2021 Published by Elsevier B.V. on behalf of Tongji University, Tongji University Press. This is an open access article under the CC BY-NC-ND license (http://***/ licenses/by-nc-nd/4.0/).

关键词： Work zone Freeway safety Real -time crash prediction machine learning Convolutional Neural Network Binary Logistic Regression

来源：评论

学校读者我要写书评

暂无评论

Urban Event Detection from Spatio-temporal IoT Sensor data Using Graph-Based machine learning

Urban Event Detection from Spatio-temporal IoT Sensor Data U...

引用

IEEE international conference on Big data and Smart Computing (BigComp)

作者： Park, Dae-Young Ko, In-Young Korea Adv Inst Sci & Technol Sch Comp Daejeon South Korea

ISBN: (纸本)9781665421973

The accurate detection and handling of urban events such as emergency incidents are critical to improve the safety and convenience of people's life in urban environments. Recently, it has become possible to detect and handle urban events in an effective manner by analyzing the detailed urban data collected from Internet of Things (IoT) sensors. Especially, some recent works investigated the use of spatio-temporal sensor data for urban event detection. However, we found there is one challenge of having less accuracy of detecting urban events as the granularity of processing data in the spatial dimension becomes finer. To meet the challenge, we propose a novel graph-based approach that analyzes geo-spatial characteristics of urban sensor data over time to keep the accuracy of detecting urban anomaly in finergrained geo-spatial scales, by identifying and exploiting regions that have abnormal urban dynamics. Through experiments using real-world urban datasets, we show our approach effectively addresses the challenge and outperforms the popular machine-learning-based urban event detection methods.

关键词： urban event detection internet of things spatio-temporal sensor dataset machine learning graph centrality indicator

来源：评论

学校读者我要写书评

暂无评论

Optimization of Federated learning Communications with Heterogeneous Quantization 22

Optimization of Federated Learning Communications with Heter...

引用

22nd IEEE international conference on Communication Technology, ICCT 2022

作者： Liu, Peng Gao, Tianli Li, Congduan School of Electronics and Communication Engineering Sun Yat-sen University Shenzhen China Guangxi Normal University Guangxi Key Lab of Multi-source Information Mining & Security Guilin China

ISBN: (纸本)9781665470674

The rapid development of machine learning in the field of artificial intelligence benefits from a large amount of training data. Due to the problems of data fragmentation and data isolation, federated learning was proposed. However, there is a communication bottleneck in the learning process of federated learning. To solve this problem, this paper adopts the method of quantization to optimize the communication of federated learning and quantifies features with different accuracy according to the feature importance. We compare the communication overhead between the communication optimization method and the non-communication optimization method, and give a theoretical explanation based on the scenario of detecting fraud in bank credit card transactions. © 2022 IEEE.

关键词： learning systems

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：