检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zheng, Jingheng Ni, Wanli Tian, Hui Gündüz, Deniz Quek, Tony Q.S. Han, Zhu The State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing100876 China The Department of Electrical and Electronic Engineering Imperial College London LondonSW7 2AZ United Kingdom Modena41100 Italy The Singapore University of Technology and Design 487372 Singapore The Department of Electronic Engineering Kyung Hee University Yongin17104 Korea Republic of The Department of Electrical and Computer Engineering University of Houston HoustonTX77004 United States The Department of Computer Science and Engineering Kyung Hee University Seoul446-701 Korea Republic of

Under the organization of the base station (BS), wireless federated learning (FL) enables collaborative model training among multiple devices. However, the BS is merely responsible for aggregating local updates during the training process, which incurs a waste of the computational resource at the BS. To tackle this issue, we propose a semi-federated learning (SemiFL) paradigm to leverage the computing capabilities of both the BS and devices for a hybrid implementation of centralized learning (CL) and FL. Specifically, each device sends both local gradients and data samples to the BS for training a shared global model. To improve communication efficiency over the same time-frequency resources, we integrate over-the-air computation for aggregation and non-orthogonal multiple access for transmission by designing a novel transceiver structure. To gain deep insights, we conduct convergence analysis by deriving a closed-form optimality gap for SemiFL and extend the result to two extra cases. In the first case, the BS uses all accumulated data samples to calculate the CL gradient, while a decreasing learning rate is adopted in the second case. Our analytical results capture the destructive effect of wireless communication and show that both FL and CL are special cases of SemiFL. Then, we formulate a non-convex problem to reduce the optimality gap by jointly optimizing the transmit power and receive beamformers. Accordingly, we propose a two-stage algorithm to solve this intractable problem, in which we provide the closed-form solutions to the beamformers. Extensive simulation results on two real-world datasets corroborate our theoretical analysis, and show that the proposed SemiFL outperforms conventional FL and achieves 3.2% accuracy gain on the MNIST dataset compared to state-of-the-art benchmarks. Copyright © 2023, The Authors. All rights reserved.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Development of Robust Vehicle Classification and Tracking for Traffic Analysis

Development of Robust Vehicle Classification and Tracking fo...

引用

IEEE International Conference on Intelligent Transportation engineering (ICITE)

作者： Sayyedjavad Ziaratnia Yiping Le Shinri Sone Noriyasu Kachi Department of Computer Science and Engineering Shibaura Institute of Technology Tokyo Japan Department of Civil Engineering Shibaura Institute of Technology Tokyo Japan Fujida Corp. Tokyo Japan Information Networking for Innovation and Design (INIAD) Toyo University Tokyo Japan

The Japanese government plans to replace the manual survey method of counting vehicles with CCTV images in order to obtain traffic volume by vehicle types automatically. The current method of manually checking vehicles and counting them as small or large vehicles is time-consuming, and there is a shortage of surveyors. Automating this process will allow for more accurate and constant monitoring of traffic volume. Existing research can be found in developing vehicle detection and tracking models using machine learning. However, these models have not yet achieved the accuracy required for practical use. This paper aims to develop a vehicle classification and tracking model with high accuracy for 7 classes of vehicle types. YOLOv5 is used for vehicle detection and classification, along with DeepSORT for vehicle tracking. The results showed most of the classes reached 95% accuracy both for classification and tracking.

关键词：

来源：评论

学校读者我要写书评

暂无评论

YOLOv8-ResCBAM: YOLOv8 Based on An Effective Attention Module for Pediatric Wrist Fracture Detection

arXiv

引用

arXiv 2024年

作者： Ju, Rui-Yang Chien, Chun-Tse Chiang, Jen-Shiun Graduate Institute of Networking and Multimedia National Taiwan University No. 1 Sec. 4 Roosevelt Rd. Taipei City106319 Taiwan Department of Electrical and Computer Engineering Tamkang University No.151 Yingzhuan Rd. Tamsui Dist. New Taipei City251301 Taiwan

Wrist trauma and even fractures occur frequently in daily life, particularly among children who account for a significant proportion of fracture cases. Before performing surgery, surgeons often request patients to undergo X-ray imaging first, and prepare for the surgery based on the analysis of the X-ray images. With the development of neural networks, You Only Look Once (YOLO) series models have been widely used in fracture detection for computer-Assisted Diagnosis, where the YOLOv8 model has obtained the satisfactory results. Applying the attention modules to neural networks is one of the effective methods to improve the model performance. This paper proposes YOLOv8-ResCBAM, which incorporates Convolutional Block Attention Module integrated with resblock (ResCBAM) into the original YOLOv8 network architecture. The experimental results on the GRAZPEDWRI-DX dataset demonstrate that the mean Average Precision calculated at Intersection over Union threshold of 0.5 (mAP 50) of the proposed model increased from 63.6% of the original YOLOv8 model to 65.8%, which achieves the state-of-the-art performance. The implementation code is available at https://***/RuiyangJu/Fracture_Detection_Improved_YOLOv8. Copyright © 2024, The Authors. All rights reserved.

关键词： Medical image processing

来源：评论

学校读者我要写书评

暂无评论

FAS-RIS: A Block-Correlation Model Analysis

arXiv

引用

arXiv 2024年

作者： Lai, Xiazhi Yao, Junteng Zhi, Kangda Wu, Tuo Morales-Jimenez, David Wong, Kai-Kit The School of Computer Science Guangdong University of Education Guangdong Guangzhou China The Faculty of Electrical Engineering and Computer Science Ningbo University Ningbo315211 China The School of Electronic Engineering and Computer Science Queen Mary University of London LondonE1 4NS United Kingdom Department of Signal Theory Networking and Communications Universidad de Granada Granada18071 Spain The Department of Electronic and Electrical Engineering University College London LondonWC1E 6BT United Kingdom The Yonsei Frontier Laboratory The School of Integrated Technology Yonsei University Seoul03722 Korea Republic of

In this correspondence, we analyze the performance of a reconfigurable intelligent surface (RIS)-aided communication system that involves a fluid antenna system (FAS)-enabled receiver. By applying the central limit theorem (CLT), we derive approximate expressions for the system outage probability when the RIS has a large number of elements. Also, we adopt the block-correlation channel model to simplify the outage probability expressions, reducing the computational complexity and shedding light on the impact of the number of ports. Numerical results validate the effectiveness of our analysis, especially in scenarios with a large number of RIS elements. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IoT-Enhanced Path Planning and Obstacle Avoidance for Autonomous Mobile Robots

IoT-Enhanced Path Planning and Obstacle Avoidance for Autono...

引用

Advancement in Technology (ICONAT), International Conference for

作者： N Selvanathan Sr Ramya C Sureshkumar R. Kiruthiga J Akshya Mani Deepak Choudhry M. Sundarrajan Department of Information Technology Sona College of Technology Salem Department of Computer Science and Engineering Dr NGP Institute of Technology Coimbatore Department of Information Technology KGiSL Institute of Technology Coimbatore Department of AI&DS Bannari Amman Institute of Technology Erode Department of Computational Intelligence SRM Institute of Science and Technology Chennai Department of Computing Technologies SRM Institute of Science and Technology Chennai Department of Networking and Communications SRM Institute of Science and Technology Chennai

ISBN: (数字)9798350354171

ISBN: (纸本)9798350354188

Path planning and obstacle avoidance play a key role in how well autonomous mobile robots work and stay safe when they move through tricky areas. Traditional path-planning methods often can’t keep up when new obstacles pop up or existing ones change, which makes them less useful in the real world. To fix these problems, the proposed research work brings in a new IoT-powered path planning and obstacle avoidance system that uses the $\mathrm{D}^{*}$ Lite and Deep Reinforcement Learning (DRL) algorithms. Our approach brings together D* Lite’s quick replanning skills with DRL’s ability to learn and adapt made even better by live data from IoT sensors. We’ve put our model through its paces in different simulated environments set up to be as tricky as real-world navigation. Quantitative analysis shows it’s a big step up from older models. Adding IoT data leads to a 30% boost in how paths are planned and cut the time robots need to get through complex routes by 20%. Plus, using DRL helps robots deal with moving obstacles 25% better, which means fewer potential crashes. This is a big step forward in the field of autonomous robotics. It gives a robust answer that would not just be reactive to the given environmental inputs but also, through proactivity, prompt possible obstacles, further contributing to smoother and safer output. As we move along with the development of our methodology, we foresee broad applications in many sectors enhancing the autonomy and efficiency of robotic systems working in dynamic and unpredictable environments.

关键词： Measurement Adaptation models Navigation Statistical analysis Heuristic algorithms Robot sensing systems Path planning Mobile robots Collision avoidance Autonomous robots

来源：评论

学校读者我要写书评

暂无评论

The Role of Blockchain, AI and IoT for Smart Road Traffic Management System 1

The Role of Blockchain, AI and IoT for Smart Road Traffic Ma...

引用

1st IEEE India Council International Subsections Conference, INDISCON 2020

作者： Sharma, Ashish Awasthi, Yogesh Kumar, Sunil College of Engineering and Computer Science Lebanese French University Department of Computer Networking Erbil Iraq College of Engineering and Computer Science Lebanese French University Department of Information Technology Erbil Iraq Amity School of Engineering and Technology Amity University Noida UP India

ISBN: (纸本)9781728187341

Nowadays vehicles are increasing on the road. Due to this, it is a challenge for society to manage traffic jams and road accidents all over the world. Artificial Intelligence (AI) such as Machine Learning (ML) algorithms are very helpful to improve the performance of the overall road safety management system. AI is used for many real-world applications to make any system be a smart system. The Smart Road Traffic Management System (SRTMS) easily recognizes the influence occurs for random changes on road safety. The SRTMS detects the unsafe driving patterns as well as convey the information to the respective authorities. The Internet of Things (IoT) is a boon technology to observe human activities in real-time. IoT devices or nodes are composed of sensors that are commonly utilized to identify and reply to electrical and other signals. Currently, Blockchain (BC) is the most trending technology to automate transactions, which means sharing or exchange of information between the IoT devices or nodes. BC technology facilitates for sharing of information on the network is decentralized, secure, persistent, anonymity, suitability and trustworthy manner. With consensus algorithms and smart contracts, Blockchain holds to manage communication among nodes without the involvement of a third-party or intermediary body. Simultaneously, AI has the ability to offer intelligent and decision-making machines similar to human beings' minds. This paper proposes the SRTMS model for solving the road accident, traffic jam and disseminate the information to all stakeholders. This proposed model is a combination of most trending technologies such as AI, BC, and IoT. This paper proposes the SRTMS model for solving the road accident, traffic jam and disseminate the information to all stakeholders. This proposed model is a combination of most trending technologies such as AI, BC, and IoT. © 2020 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Autonomous Decision-Making in Cyber-Physical Systems Using Double Deep Q-Learning and Multi-Agent Collaboration

Autonomous Decision-Making in Cyber-Physical Systems Using D...

引用

Advancement in Computation & computer Technologies (InCACCT), International Conference on

作者： Divyabarathi P Anbarasan R. Vasanthi R B. Leena Akshya J M. Sundarrajan Mani Deepak Choudhry Department of Artificial Intelligence and Data Science Bannari Amman Institute of Technology Erode Tamil Nadu India Department of Information Technology KGiSL Institute of Technology Coimbatore Tamil Nadu India Department of Computer Science and Engineering Bharathidasan Engineering College Thirupathur Tamil Nadu India Department of Electronics and Communication Engineering KGiSL Institute of Technology Coimbatore Tamil Nadu India Department of Computational Intelligence SRM Institute of Science and Technology Chennai Tamil Nadu India Department of Networking and Communications SRM Institute of Science and Technology Chennai Tamil Nadu India Department of Computing Technologies SRM Institute of Science and Technology Chennai Tamil Nadu India

ISBN: (数字)9798331543891

ISBN: (纸本)9798331543907

The autonomous decision-making capabilities of cyber-physical systems (CPS) establish key elements that optimize the management of resources while distributing tasks and maximizing system operational efficiency. Deep Q-Learning (DQN) represents one approach along with Twin Delayed Deep Deterministic Policy Gradient (TD3) and Multi-Agent Proximal Policy Optimization (PPO) that demonstrates capability in handling decision tasks. The existing models face multiple implementation challenges including inaccurate Q-value estimations weak multi-agent relationships and restricted scalability for dynamic systems. The inadequate performance of present methods for energy efficiency and latency alongside poor collaborative coordination led to the necessity of developing a new approach. The new framework DDQN-MAC takes double-deep Q-Learning and combines it with multi-agent collaboration technology to enhance decision-making abilities specific to challenging environments. The method solves overestimation bias through separate action selection and evaluation processes that use multi-agent collaboration for dynamic task distribution and resource control. The designed model demonstrates enhanced performance in multiple critical evaluation measures. Benchmarks indicate that DDQN-MAC attained a superior completion rate of 95.4% while demonstrating reduced latency performance at 12.5 ms versus the other methods DQN (18.7 ms) and Multi-Agent PPO (92.3%). The research provides fundamental groundwork to guide upcoming investigations in hybrid model development while advancing complex multi-agent system capabilities for enhanced autonomous navigational choices within variable operational environments.

关键词： Q-learning Decision making Collaboration Estimation Cyber-physical systems Dynamic scheduling Energy efficiency Resource management Vehicle dynamics Optimization

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Framework for Oversubscription Management in CPU-GPU Unified Memory

arXiv

引用

arXiv 2022年

作者： Long, Xinjian Gong, Xiangyang Zhou, Huiyang State Key Lab. of Networking and Switching Tech Beijing University of Posts and Telecommunications Beijing China Department of Electrical and Computer Engineering North Carolina State University Raleigh United States

Unified Virtual Memory (UVM) improves GPU's programmability by enabling on-demand data movement between CPU memory and GPU memory. Thanks to this emerging feature, GPUs become more ubiquitous in systems ranging from servers to data centers, and there has been an increasing trend of adopting GPUs for large-scale and general-purpose applications. However, this trend soon creates a dilemma that the limited capacity of the GPU device memory is oversubscribed by the ever-growing application working set. Oversubscription overhead becomes a major performance bottleneck for data-intensive workloads running on GPU with UVM. This paper proposes a novel intelligent framework for oversubscription management in CPU-GPU UVM. We analyze the current rule-based methods of GPU memory oversubscription with unified memory, and the current learning-based methods for other computer architectural components. We then identify the performance gap between the existing rule-based methods and the theoretical upper bound. We also identify the advantages of applying machine intelligence and the limitations of the existing learning-based methods. This paper proposes a novel intelligent framework for oversubscription management in CPU-GPU UVM. It consists of an access pattern classifier followed by a pattern-specific Transformer-based model using a novel loss function aiming for reducing page thrashing. A policy engine is designed to leverage the model's result to perform accurate page prefetching and pre-eviction. We evaluate our intelligent framework on a set of 11 memory-intensive benchmarks from popular benchmark suites. Our solution outperforms the state-of-the-art (SOTA) methods for oversubscription management, reducing the number of pages thrashed by 64.4% under 125% memory oversubscription compared to the baseline, while the SOTA method reduces the number of pages thrashed by 17.3%. Our solution achieves an average IPC improvement of 1.52X under 125% memory oversubscription, and our soluti

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Managing Service Dependency for Cloud Reliability: The Industrial Practice

Managing Service Dependency for Cloud Reliability: The Indus...

引用

IEEE International Conference on Software Reliability engineering Workshops (ISSRE Wksp)

作者： Tianyi Yang Baitong Li Jiacheng Shen Yuxin Su Yongqiang Yang Michael R. Lyu Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong SAR School of Software Engineering Sun Yat-Sen Univeristy Zhuhai China Computing and Networking Innovation Lab Huawei Cloud Shenzhen China

ISBN: (数字)9781665476799

ISBN: (纸本)9781665476805

Interactions between cloud services result in service dependencies. Evaluating and managing the cascading impacts caused by service dependencies is critical to the reliability of cloud systems. This paper summarizes the dependency types in cloud systems and demonstrates the design of the Dependency Management System (DMS), a platform for managing the service dependencies in the production cloud system. DMS features full-lifecycle support for service reliability (i.e., initial service deployment, service upgrade, proactive architectural optimization, and reactive failure mitigation) and refined characterization of the intensity of dependencies.

关键词： Conferences Production Reliability engineering Software reliability Optimization

来源：评论

学校读者我要写书评

暂无评论

Automation Intervention for Monitoring Agriculture Farming Scale: A Review

Automation Intervention for Monitoring Agriculture Farming S...

引用

International Conference on Computing and networking Technology (ICCNT)

作者： Kapil Joshi Santosh Kumar Shruti Bhatla Vishal Kaushik Harishchander Anandaram Anil Kumar Department of CSE Uttaranchal Institute of Technology Uttaranchal University Dehradun India Department of Mathematics Sharda School of Basic Sciences and Research Sharda University Greater Noida UP India Department of Computer Science and Engineering Graphic Era Hill University Dehradun Uttarakhand India School of Computer Science UPES Dehradun India Centre for Computational Engineering and Networking Amrita Vishwa Vidyapeetham Coimbatore Tamil Nadu India School of Computing DIT University Dehradun India

Agricultural data is only valuable when it has been managed successfully. As data has evolved into the fundamental ingredient in contemporary agriculture to assist farmers with vital decision, latest developments in managing data are causing Smart Agriculture to rise dramatically. With factual information obtained by sensors, significant advantages emerge with the goal of enhancing production and ecology. This type of data-driven farm management is built on information that can boost efficiency by preventing resource waste and contamination of the ecosystem. Data-driven farming, with the assistance of automated solutions using artificial intelligence techniques, lays the groundwork for potential sustainable *** article analyses the current state of sophisticated agricultural systems by rehashing each critical step, from data processing in field crops to variable interest rate implementations, so that farmers can make optimised decision making to save funds while safeguarding the natural environment and converting how meal would be generated to meet the upcoming population explosion.

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：