检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Yin, Ming Chen, Wenjing Wang, Mengdi Wang, Yu-Xiang Department of Computer Science UC Santa Barbara United States Department of Statistics and Applied Probability UC Santa Barbara United States Electrical and Electronics Engineering Department Texas A&M University United States Department of Electrical and Computer Engineering Princeton University United States

Goal-oriented Reinforcement Learning, where the agent needs to reach the goal state while simultaneously minimizing the cost, has received significant attention in real-world applications. Its theoretical formulation, stochastic shortest path (SSP), has been intensively researched in the online setting. Nevertheless, it remains understudied when such an online interaction is prohibited and only historical data is provided. In this paper, we consider the offline stochastic shortest path problem when the state space and the action space are finite. We design the simple value iteration-based algorithms for tackling both offline policy evaluation (OPE) and offline policy learning tasks. Notably, our analysis of these simple algorithms yields strong instance-dependent bounds which can imply worst-case bounds that are near-minimax optimal. We hope our study could help illuminate the fundamental statistical limits of the offline SSP problem and motivate further studies beyond the scope of current consideration. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Deep Study on Encroachment Recognition Model through W-S-N and I-O-T

A Deep Study on Encroachment Recognition Model through W-S-N...

引用

Advance Computing and Innovative Technologies in engineering (ICACITE), International Conference on

作者： Krishna Kant Dixit Satyendra Singh P. Ravi Kiran Kassem Al-Attabi Aashna Sinha Sanjeev Kumar Angadi Department of Electrical Engineering GLA University Mathura Chitkara Centre for Research and Development Chitkara University Himachal Pradesh India Department of ECE CMR College of Engineering & Technology The Islamic university Najaf Iraq School of Applied and Life Sciences Divison of Research and Innovation Uttaranchal University Dehradun Uttarakhand Computer Science and Engineering Nutan College of Engineering and Research Talegaon Dabhade Pune Maharashtra India

ISBN: (数字)9798350360165

ISBN: (纸本)9798350360172

The burgeoning challenge of Internet of Things (IoT) technology heralds’ transformative capability for agriculture, mainly in enhancing the safety and operational performance of huge-scale farmlands. This paper introduces an revolutionary IoT-based framework that synergizes Global System for Mobile Communication (GSM) and Global Positioning System (GPS) modules to forge a whole, actual-time tracking and intrusion detection answer tailored for agricultural domain names. Our proposed framework is designed to address the dual demanding situations of optimizing farm management practices while ensuring robust safety in competition to capacity intrusions. By deploying a network of brand-new sensors across farmlands, our system enables the seamless collection, evaluation, and transmission of essential environmental statistics, thereby enabling farmers to make knowledgeable alternatives all of sudden. Furthermore, the combination of GSM and GPS technology enhances the tool’s functionality to without delay alert farmers about unusual sports activities, making sure immediate reaction to protection breaches. The pilot deployment of our framework demonstrates its effectiveness in no longer best substantially improving tracking accuracy however moreover in bolstering the safety of agricultural lands towards unauthorized get right of entry to.

关键词： GSM Intrusion detection Mobile communication Sensor systems Sensors Delays Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Burned Area Mapping in Greece: Investigating the Impact of Precipitation, Pre- and Post-Processing of Sentinel-1 Data in Google Earth Engine

Unsupervised Burned Area Mapping in Greece: Investigating th...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Daniel Paluba Lorenzo Giuliano Papale Triantafyllia-Maria Perivolioti Přemysl Štych Josef Laštovička Panagiotis Kalaitzis Georgia Karadimou Elena Papageorgiou Antonios Mouratidis Department of Applied Geoinformatics and Cartography Faculty of Science Charles University Department of Civil and Computer Science Engineering Tor Vergata University of Rome School of Geology Faculty of Sciences Aristotle University of Thessaloniki Balkan Center Center for Interdisciplinary Research and Innovation Aristotle University of Thessaloniki (CIRI-AUTH) Department of Geography University of the Aegean

Wildfires are one of the most significant threats to ecosystems and are increasing in frequency globally. The aim of this study is to monitor the evolution of selected wildfires in Greece that occurred during August 2021 using Sentinel-1 SAR data and unsupervised k-means clustering in Google Earth Engine. First, changes in time series after the start of the fire and the influence of precipitation were investigated. In this study, the influence of different speckle filters and post-classification filters on clustering results was tested. The difference Normalized Burn Ratio Index (dNBR) derived from Sentinel-2 data was used as a validation dataset to assess accuracy using the F1-score, overall accuracy, omission and commission error. The best achieved F1-scores were higher than 0.70 with omission error lower than 35% in all selected areas, where the Lee speckle filter with an 11x11 kernel window size and a 2 ha post-classification filter performed the best.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Quest for Omnioculars: Embedded Visualization for Augmenting Basketball Game Viewing Experiences

arXiv

引用

arXiv 2022年

作者： Lin, Tica Chen, Zhutian Yang, Yalong Chiappalupi, Daniele Beyer, Johanna Pfister, Hanspeter John A. Paulson School of Engineering and Applied Sciences Harvard University CambridgeMA United States The Department of Computer Science Virginia Tech BlacksburgVA United States The Department of Computer Science Eth Zürich Switzerland

Sports game data is becoming increasingly complex, often consisting of multivariate data such as player performance stats, historical team records, and athletes' positional tracking information. While numerous visual analytics systems have been developed for sports analysts to derive insights, few tools target fans to improve their understanding and engagement of sports data during live games. By presenting extra data in the actual game views, embedded visualization has the potential to enhance fans' game-viewing experience. However, little is known about how to design such kinds of visualizations embedded into live games. In this work, we present a user-centered design study of developing interactive embedded visualizations for basketball fans to improve their live game-watching experiences. We first conducted a formative study to characterize basketball fans' in-game analysis behaviors and tasks. Based on our findings, we propose a design framework to inform the design of embedded visualizations based on specific data-seeking contexts. Following the design framework, we present five novel embedded visualization designs targeting five representative contexts identified by the fans, including shooting, offense, defense, player evaluation, and team comparison. We then developed Omnioculars, an interactive basketball game-viewing prototype that features the proposed embedded visualizations for fans' in-game data analysis. We evaluated Omnioculars in a simulated basketball game with basketball fans. The study results suggest that our design supports personalized in-game data analysis and enhances game understanding and engagement. © 2022, CC BY-NC-ND.

关键词： Data visualization

来源：评论

学校读者我要写书评

暂无评论

Data Fusion for Predictive Maintenance of Industrial Assets Using Digital Twin 9

Data Fusion for Predictive Maintenance of Industrial Assets ...

引用

9th International Conference on Electrical engineering, computer science and Informatics, EECSI 2022

作者： Avornu, Christian Akama, Samuel Dodzi Avornu, Eric Kofi Aveyom, Valentine School of Engineering Huzhou University Huzhou China C.K. Tedam University of Technology and Applied Sciences Department of Computer Science Navrongo Ghana Logistics and Supply Chain Ho Technical University Ho Ghana

ISBN: (数字)9786239213565

ISBN: (纸本)9781665470438

The amount of data we produce each day in every sector keeps booming. This is attributable to the accelerating growth of the Internet of Things (IoT) technologies. Data fusion has a lot of applications in the area of complex industrial process control. The concept of data fusion can also be used to measure and detect a fault in ultrahigh-precision manufacturing assets. Predictive maintenance owes much of its promise and success to effective machine learning, artificial intelligence, big data analytics, *** are widely applied to offer digital twin simulation, analysis, and prediction. The aim of this research is to study how data from multiple sensors could be analytically explored and integrated for the purpose of predictive maintenance. In other to understand the concept behind this article, public data from Microsoft Azure is used to build a predictive maintenance model using supervised learning algorithms. These algorithms can be used for predictive maintenance of machines and equipment in many manufacturing industries. © 2022 Institute of Advanced engineering and science (IAES).

关键词： Data Fusion Digital Twin IoT Technology Machine Learning Manufacturing Industry Predictive Maintenance Supervised Learning Algorithms

来源：评论

学校读者我要写书评

暂无评论

Predictive Maintenance in IoT Networks Using a Temporal Attention Mechanism Based Federated Convolutional Neural Network

Predictive Maintenance in IoT Networks Using a Temporal Atte...

引用

Innovation in Technology (ASIANCON), Asian Conference on

作者： Naga Venkata Ramakrishna Guduri R.M. Dilip Charaan Kalaiyarasi R Samuthira Pandi V D.Laxma Reddy Selvaprabhu. T Faculty of Engineering University of Technology and Applied Sciences-Al Musannah Sultanate of Oman Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai Tamil Nadu India Department of Artificial Intelligence and Data Science Rajalakshmi Institute of Technology Chennai Tamil Nadu India Centre for Advanced Wireless Integrated Technology Chennai Institute of Technology Chennai Tamil Nadu India Department of ECE MLR Institute of Technology Hyderabad India Department of CSE Dhanalakshmi College of Engineering Chennai Tamil Nadu India

ISBN: (数字)9798350354218

ISBN: (纸本)9798350354225

In the era of Industry 4.0, edge devices in industrial settings generate vast amounts of data crucial for process optimization and Predictive Maintenance (PM). Traditional cloud-based approaches for handling this data are becoming untenable due to cost, latency, and privacy issues. Federated learning emerges as a viable solution, offering localized and efficient data processing while mitigating these challenges. This paper presents a Federated Convolutional Neural Network with Temporal Attention Mechanism (FedCNN-TAM) for PM tasks in Internet of Things (IoT) networks. This model leverages temporal features in sensor data to enhance predictive accuracy. Evaluated on the CMAPSS dataset, FedCNN-TAM outperforms representative models with a significant performance gain. This model exhibits the potential for broader applications in industrial IoT, setting the stage for future research on its scalability and adaptability to multi-modal data types.

关键词： Adaptation models Temperature distribution Technological innovation Attention mechanisms Accuracy Predictive models Data models Real-time systems Convolutional neural networks Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

Diabetes Prediction Using Machine Learning

Diabetes Prediction Using Machine Learning

引用

Communication & Computing (WCONF), World Conference on

作者： Sandeep Bhatia Amit Kumar Goel Bharat Bhushan Naib Khushwant Singh Mohit Yadav Ashish Saini Department of CSE Galgotias University Greater Noida Uttar Pradesh India Department of Computer Science & Engineering U.I.E.T Maharshi Dayanand University Rohtak India Department of Applied Sciences Mathematics U.I.E.T Maharshi Dayanand University Rohtak India Department of Computer Science & Applications U.I.E.T Maharshi Dayanand University Rohtak India

Patient’s Infirmity Assumption system is a psychological gauge structure which predicts an infection in light of the information or signs went into the system and gives the specific happens. We suggest a creative replacement for the traditional methodology that resolves the drawn-out issues with expert course of action arranging. This approach is the remedy if one isn’t too certifiable and simply has to retain what kind of illness they are dealing with. It very well may be a structure that gives the client urging on the most proficient method to keep their prosperity system with everything looking good and offers a strategy to recognize disease using this expectation. The prosperity section is indispensable to the treatment of patients’ diseases, so it is furthermore a casing of help to the medical services portion that the client will be taught and helped inside the event that the individual decides not to visit a center or other facility, allowing the client to remember around their condition by entering the incidental effects and some other fitting information.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Reliable Dermatology Evaluation Benchmarks

arXiv

引用

arXiv 2023年

作者： Gröger, Fabian Lionetti, Simone Gottfrois, Philippe Gonzalez-Jimenez, Alvaro Groh, Matthew Daneshjou, Roxana Consortium, Labelling Navarini, Alexander A. Pouly, Marc Department of Biomedical Engineering University of Basel Switzerland Department of Computer Science Lucerne University of Applied Sciences and Arts Switzerland Kellogg School of Management Northwestern University United States Department of Dermatology Stanford School of Medicine United States Department of Dermatology University Hospital of Basel Switzerland Department of Dermatology University Hospital of Zurich Switzerland

Benchmark datasets for digital dermatology unwittingly contain inaccuracies that reduce trust in model performance estimates. We propose a resource-efficient data-cleaning protocol to identify issues that escaped previous curation. The protocol leverages an existing algorithmic cleaning strategy and is followed by a confirmation process terminated by an intuitive stopping criterion. Based on confirmation by multiple dermatologists, we remove irrelevant samples and near duplicates and estimate the percentage of label errors in six dermatology image datasets for model evaluation promoted by the International Skin Imaging Collaboration. Along with this paper, we publish revised file lists for each dataset which should be used for model evaluation.1 Our work paves the way for more trustworthy performance assessment in digital dermatology. Copyright © 2023, The Authors. All rights reserved.

关键词： Cleaning

来源：评论

学校读者我要写书评

暂无评论

Tuberculosis Detection from Chest Radiographs with Pre-trained Deep Learning Scheme: A Study 31

Tuberculosis Detection from Chest Radiographs with Pre-train...

引用

31st International Conference on computer Theory and Applications, ICCTA 2021

作者： Kadry, Seifedine Rajinikanth, Venkatesan Chandrasekar, A. Nandhini, A. Noroff University College Faculty of Applied Computing and Technology Kristiansand Norway St. Joseph's College of Engineering Omr Department of Electronics and Instrumentation Engineering Chennai600119 India St. Joseph's College of Engineering Omr Department of Computer Science and Engineering Chennai600119 India

ISBN: (纸本)9781665478540

In human, the abnormality in lung causes a severe respiratory problem and breathing difficulties. Tuberculosis (TB) is one of the common lung abnormality caused due a bacterium named Mycobacterium tuberculosis. TB infection will cause harsh breathing issues and untreated TB will lead to death. Detection of the TB using Chest X-ray is one of the common techniques and in this research work;TB detection using Deep-Learning-Technique (DLT) is demonstrated. The various phases of this research involve;(i) Image collection and resizing, (ii) Deep-Feature (DF) extraction with chosen technique, (iii) Classification using SoftMax classifier and (iv) Performance evaluation with existing technique. The proposed work employs a 5-fold cross validation and the best value is considered as the result. The outcome of this study confirms that, the classification accuracy achieved with ResNetl8 and K-Nearest Neighbor (ResNet+KNN) offered better outcome >97% compared to other DLT of this study. © 2021 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Impact Evaluation of Driving Style on Electric Vehicle Battery based on Field Testing Result

Impact Evaluation of Driving Style on Electric Vehicle Batte...

引用

Consumer Communications and Networking Conference, CCNC IEEE

作者： Ka Seng Chou Davide Aguiari Rita Tse Su-Kit Tang Giovanni Pau Faculty of Applied Sciences Macao Polytechnic University Macao SAR China Department of Computer Science and Engineering Alma Mater Studiorum University of Bologna Bologna IT Autonomous Robotics Research Center Technology Innovation Institute UCLA Samueli Computer Science University of California Los Angeles

Monitoring electric vehicles' battery status and forecasting their state of health is still an open challenge. To determine how and why a battery degrades over time, we have extensively monitored a Nissan Leaf's battery pack for more than one year. Collecting more than 4.5 million samples via a custom monitoring connected device to investigate how different driving behaviors affect battery aging. In addition, the best driving behaviors based on the battery's optimal temperature are revealed, including speed, acceleration and brake pedal pressure, and horsepower.

关键词： Temperature sensors Temperature measurement Aging Electric vehicles Batteries Behavioral sciences Brakes

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：