检索结果-内蒙古大学图书馆

18th Robotics: Science and Systems, RSS 2022

作者： Fulton, Michael Mehtaz, Muntaqim Sattar, Junaed Queeglay, Owen Deptartment of Computer Science & Engineering Minnesota Robotics Institute University of Minnesota Twin Cities United States Department of Electrical Engineering Stanford University United States

ISBN: (纸本)9780992374785

Autonomous underwater vehicles (AUVs) have long lagged behind other types of robots in supporting natural communication modes for human-robot interaction. Due to the limitations of the environment, most AUVs use digital displays or topside human-in-the-loop communications as their primary or only communication vectors. Natural methods for robotto-human communication such as robot "gestures" have been proposed, but never evaluated on non-simulated AUVs. In this paper, we enhance, implement and evaluate a robot-to-human communication system for AUVs called Robot Communication Via Motion (RCVM), which utilizes explicit motion phrases (kinemes) to communicate with a dive partner. We present a small pilot study that shows our implementation to be reasonably effective in person followed by a large-population study, comparing the communication effectiveness of our RCVM implementation to three baseline systems. Our results establish RCVM as an effective method of robot-to-human communication underwater and reveal the differences with more traditional communication vectors in how accurately communication is achieved at different viewpoints and types of information payloads. © 2022, MIT Press Journals. All rights reserved.

关键词： Autonomous underwater vehicles

来源：评论

学校读者我要写书评

暂无评论

MADA: Meta-Adaptive Optimizers Through Hyper-Gradient Descent 41

MADA: Meta-Adaptive Optimizers Through Hyper-Gradient Descen...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Ozkara, Kaan Karakus, Can Raman, Parameswaran Hong, Mingyi Sabach, Shoham Kveton, Branislav Cevher, Volkan Department of Electrical and Computer Engineering University of California Los Angeles United States Amazon Web Services United States Department of Electrical and Computer Engineering University of Minnesota United States Faculty of Data and Decision Sciences Technion - Israel Institute of Technology Israel LIONS IEM STI Ecole Polytechnique Fédérale de Lausanne Switzerland

Following the introduction of Adam, several novel adaptive optimizers for deep learning have been proposed. These optimizers typically excel in some tasks but may not outperform Adam uniformly across all tasks. In this work, we introduce Meta-Adaptive Optimizers (MADA), a unified optimizer framework that can generalize several known optimizers and dynamically learn the most suitable one during training. The key idea in MADA is to parameterize the space of optimizers and dynamically search through it using hypergradient descent during training. We empirically compare MADA to other popular optimizers on vision and language tasks, and find that MADA consistently outperforms Adam and other popular optimizers, and is robust against sub-optimally tuned hyper-parameters. MADA achieves a greater validation performance improvement over Adam compared to other popular optimizers during GPT-2 training and fine-tuning. We also propose AVGrad, a modification of AMSGrad that replaces the maximum operator with averaging, which is more suitable for hyper-gradient optimization. Finally, we provide a convergence analysis to show that parameterized interpolations of optimizers can improve their error bounds (up to constants), hinting at an advantage for meta-optimizers. Copyright 2024 by the author(s)

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Robust Machine Learning Technique to Classify COVID-19 Using Fusion of Texture and Vesselness of X-Ray Images

引用

computer Modeling in engineering & Sciences 2024年第2期138卷 1981-2004页

作者： Shaik Mahaboob Basha Victor Hugo Cde Albuquerque Samia Allaoua Chelloug Mohamed Abd Elaziz Shaik Hashmitha Mohisin Suhail Parvaze Pathan Department of Electronics and Communication Engineering N.B.K.R.Institute of Science&TechnologyVidyanagarAndhra Pradesh524413India Department of Teleinformatics Engineering Federal University of CearáFortaleza60455-970Brazil Information Technology Department College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh84428Saudi Arabia Department of Mathematics Faculty of ScienceZagazig UniversityZagazig44519Egypt Faculty of Computer Science and Engineering Galala UniversitySuez435611Egypt Artificial Intelligence Research Center(AIRC) College of Engineering and Information TechnologyAjman UniversityAjmanUnited Arab Emirates Department of Electrical and Computer Engineering Lebanese American UniversityByblos13-5053Lebanon Department of Avionics Indian Institute of Space Science and TechnologyThiruvananthapuramKerala695547India Department of Veterinary Clinical Sciences University of MinnesotaSt.PaulMinnesota55108USA

Manual investigation of chest radiography(CXR)images by physicians is crucial for effective decision-making in COVID-19 ***,the high demand during the pandemic necessitates auxiliary help through image analysis and machine learning *** study presents a multi-threshold-based segmentation technique to probe high pixel intensity regions in CXR images of various pathologies,including normal *** information is extracted using gray co-occurrence matrix(GLCM)-based features,while vessel-like features are obtained using Frangi,Sato,and Meijering *** learning models employing Decision Tree(DT)and RandomForest(RF)approaches are designed to categorize CXR images into common lung infections,lung opacity(LO),COVID-19,and viral pneumonia(VP).The results demonstrate that the fusion of texture and vesselbased features provides an effective ML model for aiding *** ML model validation using performance measures,including an accuracy of approximately 91.8%with an RF-based classifier,supports the usefulness of the feature set and classifier model in categorizing the four different ***,the study investigates the importance of the devised features in identifying the underlying pathology and incorporates histogrambased *** analysis reveals varying natural pixel distributions in CXR images belonging to the normal,COVID-19,LO,and VP groups,motivating the incorporation of additional features such as mean,standard deviation,skewness,and percentile based on the filtered ***,the study achieves a considerable improvement in categorizing COVID-19 from LO,with a true positive rate of 97%,further substantiating the effectiveness of the methodology implemented.

关键词： Chest radiography(CXR)image COVID-19 classifier machine learning random forest texture analysis

来源：评论

学校读者我要写书评

暂无评论

FePd Single Layer and Synthetic Antiferromagnet with Low Damping and Crystalline Perpendicular Magnetic Anisotropy on Amorphous Si/SiO2 Wafers

FePd Single Layer and Synthetic Antiferromagnet with Low Dam...

引用

2023 IEEE International Magnetic Conference - Short Papers, INTERMAG Short Papers 2023

作者： Lyu, Deyuan Shoup, Jenae E. Huang, Dingbin Wang, Xiaojia Gopman, Daniel B. Wang, Jian-Ping University of Minnesota Department of Electrical and Computer Engineering MinneapolisMN55455 United States National Institute of Standards and Technology Materials Science and Engineering Division GaithersburgMD20899 United States University of Minnesota Department of Mechanical Engineering MinneapolisMN55455 United States

ISBN: (纸本)9798350338362

(001)-textured L10-phase FePd possesses strong crystalline perpendicular magnetic anisotropy (PMA) and low Gilbert damping constant and thus, is considered a promising material choice for scalable spintronic devices. To date, however, well-textured L10-FePd(001) thin films can only be prepared on a few expensive substrates, which directly impedes their industry compatibility and application prospect. In this digest, we report the deposition of L10-FePd(001) single layer and synthetic antiferromagnet (SAF) on industry-ready Si/SiO2 wafers with large PMA, low damping, and strong antiferromagnetic interlayer exchange coupling. The microstructures of L10-FePd(001) single layer and SAF were imaged using scanning transmission electron microscopy. We observed sharp and flat interfaces between each layer as well as the heteroepitaxial growth of sample stacks, which explains the high performance of the samples. © 2023 IEEE.

关键词： Binary alloys

来源：评论

学校读者我要写书评

暂无评论

High-Lift Prediction Workshop 5: High-Order Discretization Technology Focus Group Summary

High-Lift Prediction Workshop 5: High-Order Discretization T...

引用

AIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Galbraith, Marshall C. Ruiz-Gironés, Eloi Roca, Xevi Wang, Z.J. Holst, Kevin R. Skaperdas, Vangelis Department of Aeronautics and Astronautics Massachusetts Institute of Technology CambridgeMA02139 United States Applications in Science and Engineering Barcelona Supercomputing Center - BSC Barcelona Spain Department of Aerospace Engineering University of Kansas LawrenceKS66045 United States Electrical Engineering and Computer Science University of Tennessee KnoxvilleTN37996 United States Customer Service BETA CAE Systems Thessaloniki Greece

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

This paper summarizes the High-Order Technical Focus Group (HO-TFG) submissions forthe 5th AIAA High-Lift Prediction Workshop (HLPW-5). The goal of the workshop was to assess the state-of-the-art in mesh generation and computational fluid dynamics software for predicting peak lift of transport aircraft with flaps and slats deployed. Numerous geometric build ups of the Common Research Model High-Lift (CRM-HL) variant and Reynolds number considerations served as the primary focus of the workshop, and a simple wing-body variant of the CRM-HL was used as a verification test case. The HO-TFG was tasked with generating high-order curved meshes for these complex geometries, and computing high-order solutions using both Reynolds Averaged Navier-Stokes (RANS) and scale resolving Large Eddy Simulation(LES) formulations. While participation in the HO-TFG was limited, the effort produced curved meshes on highly-complex geometric configurations and computational results from two participants. The LES results demonstrate the potential for producing results comparable to 2nd-order accurate methods, but with fewer degrees of freedom. The RANS submission for this workshop does not demonstrate a benefit over 2nd-order schemes. However, the authors will argue a lack of understanding for optimal mesh distributions for high-order methods is the cause for the limited benefits observed for the RANS results. © 2025, American institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Reynolds number

来源：评论

学校读者我要写书评

暂无评论

engineering electrode interfaces for telecom-band photodetection in MoS_(2)/Au heterostructures via sub-band light absorption

引用

Light(Science & Applications) 2023年第12期12卷 2677-2687页

作者： Chengyun Hong Saejin Oh Vu Khac Dat Sangyeon Pak SeungNam Cha Kyung-Hun Ko Gyung-Min Choi Tony Low Sang-Hyun Oh Ji-Hee Kim Department of Energy Science Sungkyunkwan UniversitySuwon 16419Republic of Korea Center for Integrated Nanostructure Physics(CINAP) Institute for Basic Science(IBS)Sungkyunkwan UniversitySuwon 16419Republic of Korea School of Electronic and Electrical Engineering Hongik UniversitySeoul 04066Republic of Korea Department of Physics Sungkyunkwan UniversitySuwon 16419Republic of Korea Department of Electrical and Computer Engineering University of MinnesotaMinneapolisMN 55455USA Department of Physics Pusan National UniversityBusan 46241Republic of Korea

Transition metal dichalcogenide(TMD)layered semiconductors possess immense potential in the design of photonic,electronic,optoelectronic,and sensor ***,the sub-bandgap light absorption of TMD in the range from near-infrared(NIR)to short-wavelength infrared(SWIR)is insufficient for applications beyond the bandgap ***,we report that the sub-bandgap photoresponse of MoS_(2)/Au heterostructures can be robustly modulated by the electrode fabrication method *** observed up to 60%sub-bandgap absorption in the MoS_(2)/Au heterostructure,which includes the hybridized interface,where the Au layer was applied via sputter *** greatly enhanced absorption of sub-bandgap light is due to the planar cavity formed by MoS_(2) and Au;as such,the absorption spectrum can be tuned by altering the thickness of the MoS_(2) *** in the SWIR wavelength range increases due to increased absorption,which means that broad wavelength detection from visible toward SWIR is *** also achieved rapid photoresponse(~150μs)and high responsivity(17 mA W^(-1))at an excitation wavelength of *** findings demonstrate a facile method for optical property modulation using metal electrode engineering and for realizing SWIR photodetection in wide-bandgap 2D materials.

关键词： MoS_(2) absorption electrode

来源：评论

学校读者我要写书评

暂无评论

Data Trading in Smart Grids: Future Research Directions

Data Trading in Smart Grids: Future Research Directions

引用

IEEE Texas Power and Energy Conference (TPEC)

作者： Soumyajit Gangopadhyay Soham Chakraborty Sarasij Das Electrical Engineering Indian Institute of Science Bangalore India Electrical and Computer Engineering University of Minnesota Minneapolis USA

The proliferation of measuring devices and sensors has resulted in a surge in smart grid (SG) data volume. However, the potential users of SG data usually face difficulty in accessing the SG data. The lack of an effective SG data exchange platform can be one of the underlying causes behind this issue. Limited accessibility of SG data can be one of the major bottlenecks towards the anticipated success of data-driven algorithms in SGs. One of the possible solutions is to establish smart grid data markets (SGDMs) for trading SG data. In this regard, recently an online survey was conducted in order to understand the stakeholders’ views on various aspects of data trading in SGs. The survey helped in identifying several obstacles including data privacy, security threat to the grid, data authenticity and money availability. These obstacles must be addressed before data trading becomes a reality in SGs. Hence, this paper proposes several directions for future research in the area of SGDMs. At first, several unique characteristics of SG data are discussed. These characteristics should be considered while addressing the challenges associated with SGDMs. Then, this paper sheds light on several areas of technical research related to data trading in SGs such as data pricing, data quality assessment, privacy pricing, privacy-preserving data trading, data value assessment, and security of data trading in SGs. In addition, this paper also identifies several areas of policy-oriented research related to SGDMs. There is a need for extensive research along the proposed research directions in order to make data trading in SGs viable in the future.

关键词： Data privacy Volume measurement Data security Data integrity Pricing Smart grids Surges

来源：评论

学校读者我要写书评

暂无评论

Grey-box System Identification of Grid-forming Inverters

Grey-box System Identification of Grid-forming Inverters

引用

IEEE Electronic Power Grid (eGrid)

作者： Samuel Helman HyeonjungTari Jung Nathan Baeckeland Deepak Ramasubramanian Sairaj Dhople Department of Electrical & Computer Engineering University of Minnesota National Renewable Energy Laboratory Electric Power Research Institute

ISBN: (数字)9798331505493

ISBN: (纸本)9798331505509

This paper demonstrates the use of grey-box system identification methods for simplifying and understanding the nonlinear power dynamics of grid-forming inverters (GFMs). The power and frequency outputs of complex high-order GFM models are fed into system identification software in order to fit them to a predetermined LTI system and learn system parameters such as (synthetic) inertia and droop constants. The same process is then run for a high-order synchronous generator model, and the outputs are fit to the same set of LTI equations. Simulation of a network of GFM inverters with diverse control architecture is also performed for the same process. The intent is threefold: first, to demonstrate the appropriateness of unified LTI models for describing the power and frequency dynamics of individual resources and connected networks, in order to facilitate analysis of larger heterogeneous networked systems; second, to discover the relationship between internal control parameters of GFMs and their externally observed values; and third, to validate that grey-box data-driven system identification techniques can be a valuable tool to discover the values of important parameters in the absence of explicit vendor models.

关键词： Linear systems Analytical models Process control Mathematical models Synchronous generators Software System identification Nonlinear dynamical systems Frequency control Grid forming

来源：评论

学校读者我要写书评

暂无评论

Revisiting the effects of the Spectre and Meltdown patches using the top-down microarchitectural method and purchasing power parity theory

引用

BenchCouncil Transactions on Benchmarks, Standards and Evaluations 2021年第1期1卷 100011-100011页

作者： Huerta, Yectli A. Lilja, David J. Scientific Computation Electrical and Computer Engineering Department Minnesota Supercomputing Institute University of Minnesota United States

Software patches are made available to fix security vulnerabilities, enhance performance, and usability. Previous works focused on measuring the performance effect of patches on benchmark runtimes. In this study, we used the Top-Down microarchitecture analysis method to understand how pipeline bottlenecks were affected by the application of the Spectre and Meltdown security patches. Bottleneck analysis makes it possible to better understand how different hardware resources are being utilized, highlighting portions of the pipeline where possible improvements could be achieved. We complement the Top-Down analysis technique with the use a normalization technique from the field of economics, purchasing power parity (PPP), to better understand the relative difference between patched and unpatched runs. In this study, we showed that security patches had an effect that was reflected on the corresponding Top-Down metrics. We showed that recent compilers are not as negatively affected as previously reported. Out of the 14 benchmarks that make up the SPEC OMP2012 suite, three had noticeable slowdowns when the patches were applied. We also found that Top-Down metrics had large relative differences when the security patches were applied, differences that standard techniques based in absolute, non-normalized, metrics failed to highlight. © 2021 The Authors

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

HISIM: Analytical Performance Modeling and Design Space Exploration of 2.5D/3D Integration for AI Computing

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Wang, Zhenyu Nalla, Pragnya Sudershan Sun, Jingbo Goksoy, A. Alper Mandal, Sumit K. Seo, Jae-Sun Chhabria, Vidya A. Zhang, Jeff Chakrabarti, Chaitali Ogras, Umit Y. Cao, Yu University of Minnesota Department of Electrical and Computer Engineering MinneapolisMN55455 United States Arizona State University School of Electrical Computer and Energy Engineering TempeAZ85287 United States Cornell Tech Department of Electrical and Computer Engineering New YorkNY10044 United States University of Wisconsin-Madison Department of Electrical and Computer Engineering MadisonWI53706 United States Indian Institute of Science Department of Computer Science and Automation Bengaluru560012 India

Monolithic designs face significant fabrication cost and data movement challenges, especially when executing complex and diverse AI models. Advanced 2.5D/3D packaging promises high bandwidth and connection density to overcome these challenges, yet it also introduces new electro-thermal constraints. This paper develops a suite of analytical performance models to enable efficient benchmarking of a 2.5D/3D heterogeneous system for energy-efficient AI computing. These models encompass various performance metrics related to computing units, network-on-chip, and network-on-package. The results are summarized into a new tool, HISIM, which is 104-106× faster than state-of-the-art AI benchmark tools. Furthermore, HISIM integrates rapid thermal simulation for the 2.5D/3D system, helping shed light on both the potential and limitations of 2.5D/3D heterogeneous integration on representative AI algorithms. The code of HISIM is available at https://***/mec-UMN/HISIM. © 1982-2012 IEEE.

关键词： Space research

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：