检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Pandey, Chetraj Ji, Anli Angryk, Rafal A. Georgoulis, Manolis K. Aydin, Berkay Dept. of Computer Science Georgia State University AtlantaGA United States Research Center for Astronomy and Applied Mathematics Academy of Athens Athens Greece

Solar flare prediction is a central problem in space weather forecasting and has captivated the attention of a wide spectrum of researchers due to recent advances in both remote sensing as well as machine learning and deep learning approaches. The experimental findings based on both machine and deep learning models reveal significant performance improvements for task specific datasets. Along with building models, the practice of deploying such models to production environments under operational settings is a more complex and often time-consuming process which is often not addressed directly in research settings. We present a set of new heuristic approaches to train and deploy an operational solar flare prediction system for ≥M1.0-class flares with two prediction modes: full-disk and active region-based. In full-disk mode, predictions are performed on full-disk line-of-sight magnetograms using deep learning models whereas in active region-based models, predictions are issued for each active region individually using multivariate time series data instances. The outputs from individual active region forecasts and full-disk predictors are combined to a final full-disk prediction result with a meta-model. We utilized an equal weighted average ensemble of two base learners’ flare probabilities as our baseline meta learner and improved the capabilities of our two base learners by training a logistic regression model. The major findings of this study are: (i) We successfully coupled two heterogeneous flare prediction models trained with different datasets and model architecture to predict a full-disk flare probability for next 24 hours, (ii) Our proposed ensembling model, i.e., logistic regression, improves on the predictive performance of two base learners and the baseline meta learner measured in terms of two widely used metrics True Skill Statistic (TSS) and Heidke Skill Score (HSS), and (iii) Our result analysis suggests that the logistic regression-based ensemble (Meta

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

A smart ocean observation system for reliable real-time measurements

A smart ocean observation system for reliable real-time meas...

引用

OCEANS

作者： Camilla Saetre Astrid Marie Skålvik Kjell-Eivind Frøysa Marie Bueie Holstad Department of Physics and Technology University of Bergen Bergen Norway Dept. of Computer Science Electrical Engineering and Mathematical Sciences Western Norway University of Applied Sciences Bergen Norway NORCE Norwegian Research Center Bergen Norway

This paper presents a research and innovation centre for a smart ocean observation system. The main goal of the centre, SFI Smart Ocean, is to enable sustainable ocean management through real-time measurements from autonomous and smart sensors. As for smart systems on-land, a smart ocean observation system of sensors requires wireless communication and a software platform gathering all required data. To ensure usability in real-time, minimize maintenance costs, and time-consuming delayed mode data analysis, it is paramount to ensure high reliability of the system. This can be achieved by on-line quality control, self-diagnostics, and self-calibration at the sensors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hands-on Active Learning Approach to Teach Artificial Intelligence/Machine Learning to Elementary and Middle School Students

Hands-on Active Learning Approach to Teach Artificial Intell...

引用

International Conference on Wireless and Optical Communications ( WOCC)

作者： Neelu Sinha Ryan F. Evans Mackenzie Carbo Department of Mathematics and Computer Science Fairleigh Dickinson University Madison NJ USA Dept. of Math and Comp Sci FDU CS Hub and Haskell Elementary School Wanaque School District Haskell NJ USA

The last decade has seen a surge in expanding access to computer science (CS) education, especially for K-12, with many states even stipulating student learning standards in CS and Computational Thinking (CT). Our 21st century K-12 students are no longer just computer users, but are now required to be computationally literate creators with proficient skills both in the concepts and practices of CS and CT. At the same time, technology continues to pervade our lives and expand at a relentless pace and all aspects of our lives are now embedded in technology surrounded by Artificial Intelligence (AI). AI in the form of Machine Learning (ML) is a key technology in a diversity of applications, where we use sensors to meaningfully perceive the world around us, analyze and organize the perceived data, and autonomously use that data to make predictions and decisions. In higher education, AI/ML courses proliferate, with many institutions now conferring degrees and certifications in these. To an extent, some high schools (grades 9–12) have started introducing these concepts in a technology class, or a robotics club, or as an after-school activity. As for middle (grades 6–8) and elementary school (grades K-5), there are very few examples of such instruction. In this paper, we present a complete framework for elementary and middle school teachers to help them prepare and incorporate AI/ML lessons in their classrooms using hands-on active learning strategies. We want to empower these teachers to impart improved learning to their students, which in turn will prepare their students to become effective thinkers, problem solvers, communicators, and gain necessary skills for high-skilled and high-demand jobs. We describe a detailed AI/ML lesson plan based on standards and framework, AI4K12 big ideas, art and science of curriculum design, active learning, and culturally responsive and inclusive pedagogy. Then we discuss our experiences in teaching the same to $\boldsymbol{4}^{\mathbf{

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ultraviolet spectrum downshifting in polymer nanocomposite films for the enhancement of optoelectronic devices 17

Ultraviolet spectrum downshifting in polymer nanocomposite f...

引用

Photonic Fiber and Crystal Devices: Advances in Materials and Innovations in Device Applications XVII 2023

作者： Darwish, Abdalla M. Sarkisov, Sergey S. Patel, Darayas N. Fedoseev, Alex Rossmanith, David Grayol, Brandon Hribljan, Anja Koplitz, Brent Rashedy, Remah El Physics and Pre-Engineering Dept. Dillard University 2601 Gentilly Blvd New OrleansLA70122 United States SSS optical Technologies 515 Sparkman Drive HuntsvilleAL35816 United States Math and Computer Science Dept Oakwood University 7000 Adventist Blvd HuntsvilleAL35896 United States Solestial Inc. TempeAZ85284 United States Chemistry Dept. Tulane University 6823 St Charles Ave New OrleansLA70118 United States f125 Putney Rd battleboroVT0530 United States

ISBN: (纸本)9781510665781

We report on polymer nanocomposite films with ultraviolet (UV) spectrum down-shifting that can be transferred using the open-air multi-beam multi-target pulsed laser deposition (MBMT-PLD) on opto-electronic devices, such as photovoltaic (PV) cells, for the enhancement of their performance. The nanocomposites were made of the space-grade polymer CORIN impregnated with nanoparticles (NPs) of rare-earth (RE) doped compound NaYF4: Eu3+. NPs had a peak of down-shifted photoluminescence (PL) at 623 nm and a PL quantum yield (PLQE) of ~ 50%. The enhancement of the coated PV cells was two-fold: (a) protection from harmful solar UV radiation and (b) the increase of the PV conversion efficiency. We describe the results of characterization of the NPs using dynamic light scattering, X-ray diffraction, and optical spectroscopy. The nanocomposite films were deposited on Silicon heterojunction (SHJ), Copper-Indium-Gallium-Selenide (CIGS) and inverted metamorphic multijunction (IMM) solar cells. The cells were exposed to 365-nm UV radiation from a light emitting diode (LED). The I-V characteristics of the cells were measured with a solar simulator using AM0 filter. The proposed films improved UV stability of all three cell types: the power degradation significantly slowed down, CIGSs (by half). The proposed films have great commercial potential, especially for the applications to space power. © 2023 SPIE. All rights reserved.

关键词： Nanocomposites

来源：评论

学校读者我要写书评

暂无评论

mathematical Model of Asset and Liability Management in the Presence of Normative and Internal Constraints

Mathematical Model of Asset and Liability Management in the ...

引用

International Conference on Control Systems, mathematical Modelling, Automation and Energy Efficiency (SUMMA)

作者： Aleksandra Zhukova Anna Flerova Aleksey Chernov Nikolay Pilnik Dept. of Mathematical Modelling of Economic Systems FRC CSC of RAS MIPT Moscow Russian Federation Dept. of Mathematical Modelling of Economic Systems FRC CSC RAS Moscow Russian Federation MIPT-School of Applied Mathematics and Computer Science Moscow Institute of Physics and Technology Moscow Russian Federation

In this paper we present the mathematical formulation of the problem of planning the Asset Liability Management if the Presence of Normative and Internal Constraints. As a nonlinear optimal control problem, it has specifics due to the state constraints. The maximum principle involves verification of regularity property in order to apply computational methods. We reformulate the problem as the problem with convex objective function and linear, but mixed system of constraints. This enables application of optimization methods to compute the optimal controls, and study the dependence of controls on the parameters that describe the external environments and the management objective.

关键词：

来源：评论

学校读者我要写书评

暂无评论

From Discrete to Continuous Convolution Layers

arXiv

引用

arXiv 2020年

作者： Shocher, Assaf Feinstein, Ben Haim, Niv Irani, Michal Dept. of Computer Science and Applied Math Weizmann Institute of Science

A basic operation in Convolutional Neural Networks (CNNs) is spatial resizing of feature maps. This is done either by strided convolution (donwscaling) or transposed convolution (upscaling). Such operations are limited to a fixed filter moving at predetermined integer steps (strides). Spatial sizes of consecutive layers are related by integer scale factors, predetermined at architectural design, and remain fixed throughout training and inference time. We propose a generalization of the common Conv-layer, from a discrete layer to a Continuous Convolution (CC) Layer. CC Layers naturally extend Conv-layers by representing the filter as a learned continuous function over sub-pixel coordinates. This allows learnable and principled resizing of feature maps, to any size, dynamically and consistently across scales. Once trained, the CC layer can be used to output any scale/size chosen at inference time. The scale can be non-integer and differ between the axes. CC gives rise to new freedoms for architectural design, such as dynamic layer shapes at inference time, or gradual architectures where the size changes by a small factor at each layer. This gives rise to many desired CNN properties, new architectural design capabilities, and useful applications. We further show that current Conv-layers suffer from inherent misalignments, which are ameliorated by CC layers. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Defensive Adversarial Training for Enhancing Robustness of ECG based User Identification*

Defensive Adversarial Training for Enhancing Robustness of E...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Hongbi Jeong Junggab Son Hyunbum Kim Kyungtae Kang Dept. Applied Artificial Intelligence Major in Bio Artificial Intelligence Hanyang University Ansan Republic of Korea Dept. Computer Science University of Nevada Las Vegas Las Vegas NV USA Dept. Embedded Systems Engineering Incheon National University Incheon Republic of Korea Dept. Applied Artificial Intelligence Hanyang University Ansan Republic of Korea

ISBN: (数字)9781665468190

ISBN: (纸本)9781665468206

Electrocardiogram (ECG) based user identification has received considerable attention with the advent of wearable devices. It provides emerging applications including personal healthcare a convenient way to authenticate users as the process can be performed at the moment the user makes contact with the device. However, a recent study discovered that injecting noise into the signal transmitted from the user to an application can effectively hinder the classification process. Many efforts have been made to deal with this noise injection attack, but most approaches have focused on noise removal. In contrast, this paper proposes Defensive Adversarial Training (DAT), which involves training a model with various noisy data to enhance the robustness of deep learning-based identification algorithms. We used two types of noise, Gaussian and Laplacian, to create noisy data. In addition, a sliding-window technique was used to effectively extract useful features and to achieve better accuracy. Our simulation results demonstrate that the proposed approach is highly robust to noise injection attacks and even against random noise. A comparative analysis with noise removal schemes also shows that the proposed DAT significantly enhances the robustness of ECG-based user identification.

关键词： Training Performance evaluation Laplace equations Wearable computers Simulation Electrocardiography Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Approximate method for helical particle trajectory reconstruction in high energy physics experiments

arXiv

引用

arXiv 2022年

作者： Topolnicki, Kacper Bold, T. M. Smoluchowski Institute of Physics Jagiellonian University Kraków30-348 Poland Agh University of Science and Technology Dept. of Physics and Applied Computer Science Krakow Poland

High energy physics experiments, in particular experiments at the LHC, require the reconstruction of charged particle trajectories. Methods of reconstructing such trajectories have been known for decades, yet the applications at High Luminosity LHC require this reconstruction to be fast enough to be suitable for online event filtering. A particle traversing the detector volume leaves signals in active detector elements from which the trajectory is reconstructed. If the detector is submerged in a uniform magnetic field that trajectory is approximately helical. Since a collision event results in the production of many particles, especially at high luminosities, the first phase of trajectory reconstruction is the formation of candidate trajectories composed of a small subset of detector measurements that are then subject of resource intensive precise track parameters estimation. In this paper, we suggest a new approach that could be used to perform this classification. The proposed procedure utilizes the z coordinate in the longitudinal direction in addition to the x, y coordinates in the plane perpendicular to the direction of the magnetic field. The suggested algorithm works equally well for helical trajectories with different proximities to the beamline which is beneficial when searching for products of particles with longer lifetimes. Copyright © 2022, The Authors. All rights reserved.

关键词： Luminance

来源：评论

学校读者我要写书评

暂无评论

Technical Debt Management: The Road Ahead for Successful Software Delivery

Technical Debt Management: The Road Ahead for Successful Sof...

引用

Software Engineering: Future of Software Engineering (ICSE-FoSE), IEEE/ACM International Conference on

作者： Paris Avgeriou Ipek Ozkaya Alexander Chatzigeorgiou Marcus Ciolkowski Neil A. Ernst Ronald J. Koontz Eltjo Poort Forrest Shull Dept. of Computing Science University of Groningen Groningen The Netherlands Software Engineering institute Carnegie Mellon University Pittsburgh PA USA Dept. of Applied informatics University of Macedonia Thessaloniki Greece QAware GmbH München Germany Department of Computer Science University of Victoria Victoria Canada The Boeing Company Mesa AZ USA CGI Rotterdam The Netherlands

Technical Debt, considered by many to be the ‘silent killer’ of software projects, has undeniably become part of the everyday vocabulary of software engineers. We know it compromises the internal quality of a system, either deliberately or inadvertently. We understand Technical Debt is not all derogatory, often serving the purpose of expediency. But, it is associated with a clear risk, especially for large and complex systems with extended service life: if we do not properly manage Technical Debt, it threatens to “bankrupt” those systems. Software engineers and organizations that develop software-intensive systems are facing an increasingly more dire future state of those systems if they do not start incorporating Technical Debt management into their day to day practice. But how? What have the wins and losses of the past decade of research and practice in managing Technical Debt taught us and where should we focus next? In this paper, we examine the state of the art in both industry and research communities in managing Technical Debt; we subsequently distill the gaps in industrial practice and the research shortcomings, and synthesize them to define and articulate a vision for what Technical Debt management looks like five years hence.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Location privacy-preserving Mobile Crowd Sensing with Anonymous Reputation 16

Location privacy-preserving Mobile Crowd Sensing with Anonym...

引用

16th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2020

作者： Mille, Arthur Karim, Lutful Almhana, Jalal Khan, Nargis Moncton University Dept. of Computer Science Canada School of ICT Seneca College of Applied Arts and Technology

ISBN: (纸本)9781728131290

In our society, the concept of sharing information is growing that lead to the emerging crowdsensing technology where a crowd of connected devices having sensors is used to sense and collect information. However, privacy is a true concern in crowdsensing technologies. Hence, this paper introduces two user's location privacy-preserving techniques based on anonymous reputation such as pseudonyms and different data collector/sender. Initially, we present very popular existing approach that uses pseudonyms to attach with the information to transmit data to the central server. Then, we introduce a simple but very effective approach DDCS, which uses different data collector and sender (i.e., one user collects information and swaps with other user that transmits collected information to the central server or platform). Finally, we introduce a hybrid approach that integrates pseudonyms with DDCS approach to achieve better privacy. Simulation results showed that hybrid approach outperforms pseudonym-based approach and DDCS in term of privacy preserving. © 2020 IEEE.

关键词： Privacy-preserving techniques

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：