检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Arora, Raman Bassily, Raef Guzmán, Cristóbal Menart, Michael Ullah, Enayat Department of Computer Science The Johns Hopkins University United States The Ohio State University United States Department of Applied Mathematics University of Twente Institute for Mathematical and Computational Engineering Pontificia Universidad Católica de Chile Chile Department of Computer Science & Engineering The Ohio State University United States

We study the problem of (ϵ, δ)-differentially private learning of linear predictors with convex losses. We provide results for two subclasses of loss functions. The first case is when the loss is smooth and non-negative but not necessarily Lipschitz (such as the squared loss). For this case, we establish an upper bound on the excess population risk of e O (kw*√nk + minn w*2 (nϵ)2/3 , √dw*2 nϵ }), where n is the number of samples, d is the dimension of the problem, and w*is the minimizer of the population risk. Apart from the dependence on w*, our bound is essentially tight in all parameters. In particular, we show a lower bound of Ω(1 √n + minn{w*4/3 (nϵ)2/3 , √dw* nϵ }). We also revisit the previously studied case of Lipschitz losses [SSTT20]. For this case, we close the gap in the existing work and show that the optimal rate is (up to log factors) Θ({kw*√nk + min nkw*√nϵk , √rankw* nϵ }), where rank is the rank of the design matrix. This im- proves over existing work in the high privacy regime. Finally, our algorithms involve a private model selection approach that we develop to enable attaining the stated rates without a-priori knowledge of w*. © 2022, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Strategies of Collaborative Filtering Recommendation Mechanism Using a Deep Learning Approach 3

Strategies of Collaborative Filtering Recommendation Mechani...

引用

3rd IEEE Eurasia Conference on IOT, Communication and engineering, ECICE 2021

作者： Chang, Chin-Chih Chang, Huan-Ming Chung Hua University Becholar Program in Applied Foreign Language and Smart Commerce Hsinchu30012 Taiwan Chung Hua University Department of Computer Science and Information Engineering Hsinchu30012 Taiwan

ISBN: (纸本)9781665445160

Nowadays, recommendation systems are widely used to help users locate the items they want. Collaborative filtering (CF) is a commonly used method for the recommendation. CF techniques use user-item ratings for prediction but suffer from the problems of data sparsity, cold start, and scalability. Though the Matrix Factorization (MF) techniques like Singular Value Decomposition (SVD) or Principal Component Analysis (PCA) overcome the above-mentioned problems, these methods are possible to deliver unmeaningful results in the condition of a low ranked approximation and denser singular vectors. In this paper, we review strategies of collaborative filtering recommendation mechanisms and propose an approach based on an autoencoder of convolutional neural network. Autoencoders are unsupervised learning methods in which neural networks are supported for the task of representation learning. We identify the user's features through learning, and then use these features to combine the collaborative filtering algorithm to recommend items. The experimental results show that the convolutional autoencoder can effectively reduce the computations when the amount of data is huge and benefited from the performance of its convolutional neural network. © 2021 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Optical scanning endoscope via a single multimode optical fiber

引用

Opto-Electronic science 2024年第3期3卷 1-32页

作者： Guangxing Wu Runze Zhu Yanqing Lu Minghui Hong Fei Xu Department of Electrical and Computer Engineering National University of Singapore4 Engineering Drive 3117576Singapore College of Engineering and Applied Sciences and Collaborative Innovation Center of Advanced Microstructures Nanjing UniversityNanjing 210023China Pen-Tung Sah Institute of Micro-Nano Science and Technology Xiamen University33 Qunxian RoadXiamen 361005China

Optical endoscopy has become an essential diagnostic and therapeutic approach in modern biomedicine for directly observing organs and tissues deep inside the human body,enabling non-invasive,rapid diagnosis and *** fiber endoscopy is highly competitive among various endoscopic imaging techniques due to its high flexibility,compact structure,excellent resolution,and resistance to electromagnetic *** the past decade,endoscopes based on a single multimode optical fiber(MMF)have attracted widespread research interest due to their potential to significantly reduce the footprint of optical fiber endoscopes and enhance imaging *** comparison with other imaging principles of MMF endoscopes,the scanning imaging method based on the wavefront shaping technique is highly developed and provides benefits including excellent imaging contrast,broad applicability to complex imaging scenarios,and good compatibility with various well-established scanning imaging *** this review,various technical routes to achieve light focusing through MMF and procedures to conduct the scanning imaging of MMF endoscopes are *** advancements in imaging performance enhancements,integrations of various imaging modalities with MMF scanning endoscopes,and applications are *** specific to this endoscopic imaging technology are analyzed,and potential remedies and avenues for future developments are discussed.

关键词： multimode optical fiber endoscope scanning imaging focusing wavefront shaping

来源：评论

学校读者我要写书评

暂无评论

Routing Single Photons from a Trapped Ion Using a Photonic Integrated Circuit

引用

Physical Review applied 2023年第3期19卷 034001-034001页

作者： Uday Saha James D. Siverns John Hannegan Mihika Prabhu Qudsia Quraishi Dirk Englund Edo Waks Department of Electrical and Computer Engineering University of Maryland College Park Maryland 20742 USA Institute for Research in Electronics and Applied Physics (IREAP) University of Maryland College Park Maryland 20742 USA Department of Physics University of Maryland College Park Maryland 20742 USA Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA United States Army Research Laboratory Adelphi Maryland 20783 USA Joint Quantum Institute (JQI) University of Maryland College Park Maryland 20742 USA

Trapped ions are promising candidates for nodes of a scalable quantum network due to their long-lived qubit coherence times and high-fidelity single- and two-qubit gates. Future quantum networks based on trapped ions will require a scalable way to route photons between different nodes. Photonic integrated circuits from fabrication foundries provide a compact solution to this problem. However, these circuits typically operate at telecommunication wavelengths that are incompatible with the strong dipole emissions of trapped ions. In this work, we demonstrate the routing of single photons from a trapped ion using a photonic integrated circuit. We employ quantum frequency conversion to match the emission of the ion to the operating wavelength of a foundry-fabricated silicon nitride photonic integrated circuit, achieving a total transmission of 31.0% ± 0.9% through the device. Using programmable phase shifters, we switch the single photons between the output channels of the circuit and demonstrate a 50:50 beam splitting condition. These results constitute an important step towards programmable routing and entanglement distribution in large-scale quantum networks and distributed quantum computers.

关键词： Integrated optics Optical quantum information processing Quantum communication Quantum information with trapped ions Quantum interconnects Quantum networks Trapped ions

来源：评论

学校读者我要写书评

暂无评论

Arabic Documents Clustering using Bond Energy Algorithm and Genetic Algorithm

Arabic Documents Clustering using Bond Energy Algorithm and ...

引用

International Conference on Information and Communication Systems (ICICS)

作者： Hadeel Alazzam Orieb AbuAlghanam Abdulsalam Alsmady Esra’ a Alhenawi Department of Intelligence Systems Al-Balqa Applied University Al-Salt Jordan Networks and Information Security Alahliyya Amman University Amman Jordan Department of Computer Engineering Jordan University of Science and Technology Irbid Jordan Department of Computer Science University of Jordan Amman Jordan

Text clustering is a popular data mining process used in data indexing and information retrieval. However, Existing clustering techniques suffer from several shortcomings such as sensitivity to the initial value, slow convergence, etc... In this paper, the Bond Energy Algorithm (BEA) has been used for clustering Arabic documents. The bond energy algorithm has been modified with the help of a Genetic algorithm to enhance the accuracy of the results. The efficiency of modified BEA has been compared to BEA baseline and K-means clustering in terms of precision, recall, and F-Score. The results show that Modified BEA can be used for clustering Arabic Documents, and the efficiency of the algorithm can be enhanced in the future to give a better result.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Machine Learning Based Framework for Bridge Condition Analysis

A Novel Machine Learning Based Framework for Bridge Conditio...

引用

IEEE International Conference on Big Data

作者： Mohammad Masum Nafisa Anjum Md Jobair Hossain Faruk Hossain Shahriar Nazmus Sakib Maria Valero M.A. Karim Akond Rahman Fan Wu Alfredo Cuzzocrea Department of Applied Data Science San Jose State University San Jose USA Department of Information Technology Kennesaw State University USA Department of Software Engineering and Game Development Kennesaw State University USA Department of Civil and Environmental Engineering Kennesaw State University USA Department of Computer Science and Software Engineering Auburn University USA Department of Computer Science Tuskegee University USA iDEA Lab University of Calabria Rende Italy

ISBN: (纸本)9781665480468

Bridges play a vital part in the transportation system by ensuring the connectedness of transportation systems, which is critical for a country’s social and economic prosperity by offering daily mobility to the people. However, according to the American Society of Civil Engineers (ASCE 2017), many U.S. bridges are in critical condition, raising safety issues, with 9.1 and 13.6 percent of the country’s 614,387 bridges, respectively, structurally defective, and functionally obsolete. Every day, 178 million people traverse these structurally defective bridges. Furthermore, the average annual failure rate is expected to be between 87 and 222. Bridge breakdowns have disastrous repercussions, and in many cases, result in death. While bridge authorities strive to improve bridge conditions, budget limits make it difficult to make cost-effective maintenance decisions. Bridge authorities distribute limited repair resources based on projected future bridge conditions. As a result, building a data-driven, autonomous, and effective bridge condition prediction model is critical for improving maintenance decision-making. In this paper, we present a novel bridge condition prediction framework using advanced Machine Learning (ML) algorithms on the National Bridge Inventory (NBI) dataset. The framework consists of two stages, where the most informative features from the NBI dataset are selected using the Recursive Feature Elimination process and in the 2 nd step, ML classifiers are applied to the selected features for bridge condition prediction. The experimental results show that the proposed framework can effectively predict bridge conditions by producing highly accurate results in terms of accuracy, precision, recall, and f1-score.

关键词： Bridges Decision making Transportation Machine learning Maintenance engineering Predictive models Big Data

来源：评论

学校读者我要写书评

暂无评论

Enhancing Survival Prediction after Venoarterial Extracorporeal Membrane Oxygenation Using Machine Learning

引用

ASAIO Journal 2025年 2025 Jun 2页

作者： Leng, Albert Bachina, Preetham Liu, Olivia Shou, Benjamin Racz, Charles Giliver, David A. Shpitser, Ilya Whitman, Glenn J. R. Cho, Sung-Min Division of Cardiac Surgery Department of Surgery Heart and Vascular Institute Johns Hopkins University School of Medicine BaltimoreMD United States Department of Radiology Stanford University Palo AltoCA United States Department of Applied Math Columbia University New York CityNY United States Department of Computer Science Johns Hopkins University BaltimoreMD United States Divisions of Neurosciences Critical Care and Cardiac Surgery Department of Neurology Surgery Anesthesiology and Critical Care Medicine and Neurosurgery Johns Hopkins University School of Medicine BaltimoreMD United States

In-hospital mortality after venoarterial extracorporeal membrane oxygenation (VA-ECMO) remains high. This study compared the performance of the Survival after Venoarterial ECMO (SAVE) score with machine learning (ML) models incorporating rich electronic medical record data to evaluate survival for patients on VA-ECMO support. We retrospectively reviewed adults undergoing VA-ECMO (2016-2022) at a single tertiary care center. The CatBoost algorithm was trained using leave-one-out cross-validation (LOOCV) on 74 extracted vital signs, laboratory values, and ventilator settings. Shapley Additive Explanations (SHAP) was used to identify key predictive features for logistic regression. Overall, 194 VA-ECMO patients (median age = 58 years, 36.6% female) were included, with 133 (69%) experiencing mortality. The SAVE score was compared to two predictive models: a pre-ECMO model (≤ 24 hours before cannulation) and an on-ECMO model (including up to the first 48 hours of ECMO). The LOOCV area under the receiver-operator characteristics curves (AUC) for the SAVE score, pre-ECMO, and on-ECMO models was 0.73, 0.77, and 0.83, respectively. Logistic regressions using ML-identified variables showed stepwise AUC improvements: 0.82 (pre-ECMO), 0.86 (on-ECMO), and 0.89 (combined). A novel, interpretable ML model predicted survival for VA-ECMO patients with accuracy comparable to the SAVE score. Incorporating on-ECMO variables significantly increased predictive performance and revealed novel variables associated with survival. © 2025 Lippincott Williams and Wilkins. All rights reserved.

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

Four-band effective square lattice model for Bernal-stacked bilayer graphene

arXiv

引用

arXiv 2024年

作者： Chen, Szu-Chao Mreńca-Kolasińska, Alina Liu, Ming-Hao Department of Electro-Optical Engineering National Formosa University Yunlin Taiwan AGH University Faculty of Physics and Applied Computer Science al. Mickiewicza 30 Krakow30-059 Poland National Cheng Kung University Tainan70101 Taiwan

Bernal-stacked bilayer graphene (BLG) provides an ideal basis for gate-controlled, and free of etching, electronic devices. Theoretical modeling of realistic devices is an essential part of research, however, simulations of large-scale BLG devices continue to be extremely challenging. Micrometer-sized systems are predominantly beyond the reach of the commonly used atomistic tight-binding method, while other numerical approaches based on the two dimensional Dirac equation are not straightforward to conduct due to the fermion doubling problem. Here we present an approach based on the continuum model, unharmed by the fermion doubling. The discretization of the BLG continuum Hamiltonian leads to an effective four-band model, with both valleys built-in. We demonstrate its performance with realistic, large-scale systems, and obtain results consistent with experiments and with the tight-binding model, over a broad range of magnetic field. Copyright © 2024, The Authors. All rights reserved.

关键词： Etching

来源：评论

学校读者我要写书评

暂无评论

Updating high average-utility itemsets with pre-large concept

Updating high average-utility itemsets with pre-large concep...

引用

作者： Wu, Jimmy Ming-Tai Teng, Qian Lin, Jerry Chun-Wei Yun, Unil Chen, Hsing-Chung College of Computer Science and Engineering Shandong University of Science and Technology Qingdao China Department of Computer Science Electrical Engineering and Mathematical Sciences Western Norway University of Applied Sciences Bergen Norway Department of Computer Engineering Sejong University Seoul Korea Republic of Department of Computer Science and Information Engineering Asia University Taichung Taiwan

HAUIM (High Average-Utility Itemset Mining) is a variation of HUIM (High-Utility Itemset Mining) that provides a reliable measure to reveal utility patterns in light of the length of the mined pattern. Several works have been studied to improve mining efficiency by designing multiple pruning strategies and efficient frameworks, but fewer studies have centered on the sophisticated database maintenance algorithm. Existing works still have to rescan the databases multiple times when it is necessary. We first use the pre-large principle in this paper to efficiently update the newly discovered HAUIs. For further updates and maintenance on the basis of the two thresholds, the Pre-large Average Utility Itemset (PAUI) can be maintained to increase the mining performance. Experiments will then be performed to compare the batch model, the Fast-Updated (FUP)-based model, and the Apriori-like HAUIM (APHAUIM) model designed in respect of the number of maintenance patterns, scalability, runtime, and memory usage. © 2020 - IOS Press and the authors. All rights reserved.

关键词： Maintenance

来源：评论

学校读者我要写书评

暂无评论

Forex exchange using big data analytics 1

Forex exchange using big data analytics

引用

1st International Conference on Advances in Computational science and engineering, ICACSE 2020

作者： Sadhasivam, Jayakumar Arun, M. Deepa, R. Muthukumaran, V. Lokesh Kumar, R. Prasanna Kumar, R.B. Vellore Institute of Technology Vellore India School of Computing Kalasalingam Academy of Research and Education Tamilnadu 626128 India Telangana Hyderabad India Department of Mathematics School of Applied Science REVA University Bengaluru64 India Department of Analytics School of Computer Science and Engineering Vellore Institute of Technology Vellore632 014 India Department of Computer Science REVA University Bengaluru64 India

Analysis and Prediction of forex has gained immense value in today's economy. The stock price prediction is a difficult process owing to the irregularities in stock prices. Every trader wants to know if the pattern has been repeated in past to know what the possible output of the current situation will be. The primary objective is to propose a methodology that will use a historical dataset and provide a more accurate prediction on stock price. In this paper, we will be using machine learning pattern recognition algorithm on forex tick dataset. The learned model then will produce pattern from the given dataset and on the pattern of increasing or decreasing, the buyer will initiate a buy or sell the stock respectively. We will use python coding to execute the algorithm in jupyter notebook. Matplot library will help us to perform graphing in the process and Numpy will be helpful in doing statistical analysis of data. © Published under licence by IOP Publishing Ltd.

关键词： Python

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：