检索结果-内蒙古大学图书馆

Research on Miniaturization Trend of ChatGPT Technology Model

Journal of Artificial Intelligence and Technology 2023年第3期3卷 95-99页

作者： Xuemei Shi Xiaoguang Du Xiedong Song Computer College of Huainan Normal University HuainanChina College of Computing and Information Technologies National UniversityPhilippinesPhilippines Biological Engineering of Huainan Normal University HuainanChina Department of Electronic Information Yantai City Vocational College of Science and TechnologyYantaiChina

Miniaturization and micro-miniaturization are trends in technology models,such as *** trends have the potential to enhance the practicality and professionalism of the model,as well as making them more widely ***,more individuals and organizations can leverage these technologies,and their impacts can be ***,miniaturization and micro-miniaturization can decrease the size of the model and the computing resources required,thus resulting the widespread use and development of artificial intelligence ***,they can boost the speed of model operation and training efficiency,thereby improving the practicality and efficacy of ***,this trend will have a profound impact on diverse fields,including scientific research,education,coaching,medical care,and daily life.

关键词： class ChatGPT model miniaturization practicality

来源：评论

学校读者我要写书评

暂无评论

Biological Signals for Diagnosing Sleep Stages Using Machine Learning Models 28

Biological Signals for Diagnosing Sleep Stages Using Machine...

引用

28th International computer Conference, computer Society of Iran, CSICC 2023

作者： Mostafaei, Sahar Hassanzadeh Tanha, Jafar Sharafkhaneh, Amir Agrawal, Ritwick Mostafaei, Zohair Hassanzadeh University of Tabriz Faculty of Electrical and Computer Engineering Tabriz Iran Baylor College of Medicine Section of Pulmonary Critical Care and Sleep Medicine Department of Medicine HoustonTX United States Islamic Azad University of Tabriz Branch Department of Computer Engineering Tabriz Iran

ISBN: (纸本)9798350338195

Sleep quality is important for health and can prevent diseases. Polysomnography is a standard scientific and clinical tool to evaluate sleep. Sleep staging is one of the main tasks in the sleep study that characterizes sleep cycles. In recent years, many studies are conducted using machine learning approaches to classify sleep stages. These studies can improve the accuracy and speed of the classification task, but most of them have not considered the performance of minority sleep classes. Since the number of samples in the sleep classes varies greatly, it can be considered as imbalanced data. In this study, we propose an ensemble method to handle class imbalance problem in the sleep staging task. For this purpose, we select nine biomedical signals including two EEG channels, two EOG channels, EMG, ECG, Abdominal, Thorax, and Airflow from the Sleep Heart Health Study (SHHS1) dataset. Then we use an ensemble of data-level resampling methods to rebalance the data space of sleep classes. Finally, we employ different machine learning algorithms to classify sleep stages. To evaluate the performance of the proposed method, in addition to general metrics, we use various measures such as Geometric mean (G-mean) and Matthew's Correlation Coefficient (MCC), which are proper metrics for the imbalanced data classification. The results of the developed method show that it achieves high accuracies of 0.9727, 0.9410, 0.8816, and 0.8725 for two, three, five, and six sleep classes, respectively. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

pFedES: Generalized Proxy Feature Extractor Sharing for Model Heterogeneous Personalized Federated Learning 39

pFedES: Generalized Proxy Feature Extractor Sharing for Mode...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

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

ISBN: (纸本)157735897X

Federated learning (FL), as a privacy-preserving collaborative machine learning paradigm, has attracted significant interest from industry and academia. To allow each data owner (FL client) to train a heterogeneous and personalized local model based on its local data distribution, system resources and requirements on model structure, the field of model-heterogeneous personalized federated learning (MHPFL) has emerged. Existing MHPFL approaches either rely on the availability of a public dataset with special characteristics to facilitate knowledge transfer, incur high computational and communication costs, or face potential model leakage risks. To address these limitations, we propose a model-heterogeneous personalized Federated learning approach based on generalized proxy feature Extractor Sharing (pFedES) for supervised image classification tasks. (1) We devise a shared small proxy homogeneous feature extractor before each client’s heterogeneous local model. (2) Clients train them via the proposed iterative learning to enable the exchange of global generalized knowledge and local personalized knowledge. (3) The small proxy local homogeneous extractors produced after local training are uploaded to the server for aggregation to facilitate knowledge fusion across clients. We theoretically prove pFedES converges with a non-convex convergence rate O(1/T). Experiments on 3 benchmark datasets against 9 baselines demonstrate that pFedES performs state-of-the-art model accuracy while maintaining efficient communication and computation. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

High-bandwidth Current Shunt with Extremely Low Stray Magnetic Field Based on Thin-film Resistors

引用

IEEE Transactions on Magnetics 2025年

作者： Xu, Jianzhi Yi, Lifang Lee, Woongkul Moon, Jinyeong FAMU-FSU College of Engineering Electrical and Computer Engineering Department TallahasseeFL32312 United States Apple Inc San DiegoCA92121 United States Purdue University Electrical and Computer Engineering Department West LafayetteIN47907 United States

The use of wide bandgap devices introduces higher-frequency components into power electronics due to increased switching speeds. To achieve reliable measurement of these high-frequency components in power electronics, high-bandwidth voltage and current probes are essential. While 1GHz bandwidth voltage probes are commercially available, current probes are typically limited to below 200MHz. A resistive current sensor, often called a current shunt, converts magnetic field measurement into electric potential measurement, enabling us to use well-established GHz-level voltage probes. The overall structure must have an extremely low level of stray magnetic field to minimize parasitic inductance and hence to increase the measurement bandwidth. This paper proposes a novel coaxial structure based on thin-film resistors for current shunts. ANSYS Q3D and HFSS simulations demonstrate that this structure effectively cancels magnetic fields along various axes, leading to nearly nonexistent stray fields outside the structure and significantly minimizing parasitic inductance. Experimental results from a prototype with a bandwidth of up to 2.063GHz validate the proposed design. © 1965-2012 IEEE.

关键词： Bandwidth Converter Current Field High Loss Magnetic Measurement Power Resistive Resistor Shunt Stray

来源：评论

学校读者我要写书评

暂无评论

Design of Surgical Arm Robot Based on Cloud computing 5

Design of Surgical Arm Robot Based on Cloud Computing

引用

5th International Conference on engineering Technology and its Applications, IICETA 2022

作者： Ali, Bameer H Mohammedali, Mohammed Abdulmahdi Abdul-Rahaim, Laith Ali Al-Kharsan, Ibrahim H. Najaf Iraq College of Health and Medical Techniques Kufa Iraq College of Engineering University of Babylon Electrical Engineering Department Babil Iraq The Islamic University College of Technical Engineering Computer Technical Engineering Department Najaf Iraq

ISBN: (纸本)9781665472159

The controller of the robotic system has seen a great introduction to the network that uses digital and intelligent technology, primarily in recent decades. This paper presents a cloud computing controller for a surgical robotic arm system. It performs operations that simulate surgical operations. 5 Dof arm robot controlled remotely using a soft system-based cloud computing. The controller performs by graphical design interface based on word-press and Remote X.Y. cloud system. This web page is built by word press program and introduces the proposed work. Cloud computing performs by using the Remote X.Y. app. The ESP 32 is used as a hardware controller and the interface between the surgical robot and the internet and cloud server. © 2022 IEEE.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

Enhanced QoS Performance with Throughput Aware Effective Communication in VANETs 6

Enhanced QoS Performance with Throughput Aware Effective Com...

引用

6th Iraqi International Conference on engineering Technology and its Applications, IICETA 2023

作者： Khodayer Al-Dulaimi, Aymen Mohammed Alkhazraji, Nejood Mlket Almutoki, Sabah Mohammed Alkhayyat, Ahmed H. R. Abbas, F.H. Al-Farahidi University College of Engineering Techniques Department of Communication Technical Engineering Baghdad Iraq Ja'afar AI-Sadiq University College of Information Technology Imam Department of Computer Technical Engineering Al-Muthanna Iraq Al Furat Al Awsat Technical University Technical Institute of Samawa Department of Electrical Al-Muthanna Iraq Islamic University College of Technical Engineering Department of Computer Technical Engineering Najaf Iraq Al-Mustaqbal University Al-Mustaqbal University College Medical Laboratories Techniques Department Hillah Iraq

ISBN: (纸本)9798350303339

Vehicular Ad-Hoc Network (VANET) is a highly trending and developing technology for the future world where in all the place highly intelligent communication is required. In VANETs the intelligently information is transmitted from one place to another with the base transmission model such as vehicle based and as well infrastructure-based communication. Unfortunately, wireless transmission standard uses IEEE802.11p and ground way communication model which can easily get affected by the external environment that results in the increase of packet loss and delay during communication. For that purpose, in this paper Enhanced QoS Performance with Throughput Aware Effective Communication (EQPTAE) is developed that mainly concerned with effective network model creation, collision detection and collision avoidance. Hence vehicle perform multi-hop communication in a highly frequent manner where a greater number of connection loss occur here with the help of effective congestion control communication becomes highly traffic free which leads to enhance the communication quality and throughput utility at each instant of time. In order to create this network in practical way the software called NS2 is used which is highly suitable for vehicular communication and as well the parameters which are used for the performance analysis are packet delivery ratio, end-to-end delay, routing overhead, and energy efficiency. To perform comparative analysis the earlier researches which are used in this article are ICFDB and the RSUCI. From the results it is understood that the proposed EQPTAE achieves maximum delivery ratio and efficiency and as well minimum delay and overhead when compared with the earlier methods. © 2023 IEEE.

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

High CRI white light LED using laser sintered surface PiG composite of YAG:Ce and CASN phosphors

High CRI white light LED using laser sintered surface PiG co...

引用

Optical Devices and Materials for Solar Energy and Solid-state Lighting, SOLED 2022

作者： Ali, Javed Madsen, Christi Department of Electrical and Computer Engineering Texas A&M University College StationTX77843 United States

High CRI white light using surface PiG fabricated from a mixture of YAG:Ce and CASN phosphors and borosilicate glass powder is demonstrated. Surface PiG is fabricated using pulsed CO2 laser sintering on borosilicate g... 详细信息

ISBN: (纸本)9781957171166

关键词： Phosphors

来源：评论

学校读者我要写书评

暂无评论

Hybrid Semantic/Bit Communication Based Networking Problem Optimization

Hybrid Semantic/Bit Communication Based Networking Problem O...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Le Xia Yao Sun Dusit Niyato Lan Zhang Lei Zhang Muhammad Ali Imran James Watt School of Engineering University of Glasgow Glasgow UK College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering Clemson University USA

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

This paper jointly investigates user association (UA), mode selection (MS), and bandwidth allocation (BA) problems in a novel and practical next-generation cellular network where two modes of semantic communication (SemCom) and conventional bit communication (BitCom) coexist, namely hybrid semantic/bit communication network (HSB-Net). Concretely, we first identify a unified performance metric of message throughput for both SemCom and BitCom links. Next, we comprehensively develop a knowledge matching-aware two-stage tandem packet queuing model and theoretically derive the average packet loss ratio and queuing latency. Combined with several practical constraints, we then formulate a joint optimization problem for UA, MS, and BA to maximize the overall message throughput of HSB-Net. Afterward, we propose an optimal resource management strategy by employing a Lagrange primal-dual method and devising a preference list-based heuristic algorithm. Finally, numerical results validate the performance superiority of our proposed strategy compared with different benchmarks.

关键词： Heuristic algorithms Packet loss Benchmark testing Throughput Semantic communication Resource management Global communication Optimization Queueing analysis Next generation networking

来源：评论

学校读者我要写书评

暂无评论

ContrastMask: A Novel Perturbation-Based Method for Explaining Network Intrusion Detection

ContrastMask: A Novel Perturbation-Based Method for Explaini...

引用

2024 IEEE International Conference on E-Health Networking, Application and Services, HealthCom 2024

作者： Zhu, Shuai Wang, Di Mao, Shiwen Wang, Xuyu Guangzhou Institute of Technology Xidian University Guangzhou China School of Computer Science and Technology Xidian University Xi’an China Department of Electrical and Computer Engineering Auburn University AuburnAL United States Knight Foundation School of Computing & Information Sciences Florida International University MiamiFL United States

ISBN: (纸本)9798350350548

Recently, there has been a surge in cyberattacks targeting the Internet of Health Things (IoHT), increasing the urgency for advancing network intrusion detection systems (IDS). Machine learning techniques, especially deep neural networks (DNNs), are demonstrating potential in improving the precision of detection methods. Despite their advantages, the complexity of DNNs can obscure their decision-making process, impacting their acceptance in security-critical environments. To improve the transparency of DNN models, we propose a novel post-hoc interpretation method that applies a perturbation-based approach with an optimized mask applied to an autoencoder model for IDS. More importantly, we leverage contrastive learning to maintain perturbed samples within the original feature space, reducing the risk of misclassification due to sample drift and ensuring a clear interpretation of the mask. We validate our approach using the NSL-KDD and UNSW15 datasets, showing that it provides clearer and more robust explanations compared to existing methods. This enhancement in interpretability is pivotal for healthcare cybersecurity experts to gain insights into the decision-making processes of black-box models. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Left and Right Lung-Specific Method of Segmentation in Chest Radiographs

Left and Right Lung-Specific Method of Segmentation in Chest...

引用

Applications of Machine Learning 2024

作者： Marsh, John J. Narayanan, Barath Narayanan Hardie, Russell C. Electrical and Computer Engineering Department University of Dayton 300 College Park DaytonOH45469 United States Sensor and Software Systems Division University of Dayton Research Institute 300 College Park DaytonOH45469 United States

ISBN: (纸本)9781510679368

Lung segmentation supports essential functionality within the realm of computer-aided detection and diagnosis using chest radiographs. In this research, we present a novel bifurcation approach of segmenting the lungs by separating them down the spinal column and having separate networks for the right and the left lungs respectively. Results from the right lung and left lung networks are then merged to form the overall lung. We utilize DeepLabV3+ network with ResNet50 as the backbone for both left and right lung networks. Results are presented for publicly available datasets such as Shenzhen Dataset and Japanese Society of Radiological Technology (JSRT) dataset. Our proposed bifurcation approach achieved an overall accuracy of 98.8% and an IoU (Intersection over Union) of 0.977 for a set of 100 cases in Shenzhen dataset. We conducted an additional robustness study of this method by training and testing on an independent dataset utilizing a hold-out methodology. We utilize a private dataset for the training and testing occurs on an independent JSRT dataset comprising 140 cases and our algorithm achieved an overall IoU of 0.945 thereby demonstrating its efficacy against other whole lung models and setting a new benchmark for future research works. © 2024 SPIE.

关键词： Image segmentation

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：