检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

4,353 篇 会议
68 篇 期刊文献
14 册 图书

馆藏范围

4,435 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

2,489 篇 工学
- 2,305 篇 计算机科学与技术...
- 1,055 篇 软件工程
- 492 篇 电气工程
- 430 篇 信息与通信工程
- 194 篇 电子科学与技术（可...
- 121 篇 控制科学与工程
- 102 篇 网络空间安全
- 89 篇 动力工程及工程热...
- 73 篇 生物工程
- 48 篇 机械工程
- 47 篇 生物医学工程（可授...
- 41 篇 光学工程
- 34 篇 建筑学
- 30 篇 材料科学与工程（可...
- 22 篇 仪器科学与技术
- 22 篇 土木工程
- 22 篇 化学工程与技术
604 篇 理学
- 427 篇 数学
- 97 篇 物理学
- 83 篇 生物学
- 67 篇 系统科学
- 61 篇 统计学（可授理学、...
- 23 篇 化学
339 篇 管理学
- 253 篇 管理科学与工程(可...
- 139 篇 工商管理
- 100 篇 图书情报与档案管...
61 篇 医学
- 45 篇 临床医学
- 38 篇 基础医学(可授医学...
40 篇 经济学
- 40 篇 应用经济学
38 篇 法学
- 33 篇 社会学
11 篇 教育学
11 篇 农学
6 篇 文学
1 篇 艺术学

主题

1,186 篇 computer archite...
446 篇 hardware
376 篇 high performance...
287 篇 concurrent compu...
279 篇 computational mo...
237 篇 application soft...
231 篇 parallel process...
217 篇 computer science
213 篇 distributed comp...
187 篇 costs
186 篇 bandwidth
179 篇 field programmab...
173 篇 delay
166 篇 throughput
159 篇 cloud computing
156 篇 grid computing
140 篇 computer network...
138 篇 resource managem...
133 篇 laboratories
123 篇 scalability

机构

16 篇 university of ch...
12 篇 school of comput...
12 篇 institute of com...
8 篇 department of co...
8 篇 georgia inst tec...
8 篇 univ chicago dep...
8 篇 barcelona superc...
8 篇 school of comput...
8 篇 carnegie mellon ...
8 篇 mathematics and ...
8 篇 intel corporatio...
7 篇 department of el...
7 篇 college of compu...
7 篇 univ illinois ur...
7 篇 computer systems...
7 篇 department of co...
7 篇 mathematics and ...
7 篇 intel corp santa...
7 篇 univ fed rio gra...
6 篇 univ toronto on

作者

13 篇 navaux philippe ...
11 篇 d.k. panda
9 篇 viktor k. prasan...
9 篇 prasanna viktor ...
9 篇 mutlu onur
9 篇 i. foster
8 篇 dhabaleswar k. p...
8 篇 dongarra jack
8 篇 guedes dorgival
7 篇 borin edson
7 篇 cristal adrian
7 篇 chong frederic t...
7 篇 loh gabriel h.
7 篇 kim nam sung
7 篇 zhou huiyang
7 篇 panda dhabaleswa...
7 篇 magoules frederi...
7 篇 ferreira renato
7 篇 xiaowei li
7 篇 buyya rajkumar

语言

4,403 篇 英文
23 篇 其他
10 篇 中文
2 篇 葡萄牙文
1 篇 法文

检索条件"任意字段=16th Symposium on Computer Architecture and High Performance Computing"

共 4435 条记录，以下是241-250 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

Decentralized in-order execution of a sequential task-based code for shared-memory architectures 36

Decentralized in-order execution of a sequential task-based ...

引用

36th IEEE International Parallel and Distributed Processing symposium (IEEE IPDPS)

作者： Castes, Charly Agullo, Emmanuel Aumage, Olivier Saillard, Emmanuelle Inria LaBRI Bordeaux France Ecole Polytech Fed Lausanne Lausanne Switzerland

ISBN: (纸本)9781665497473

the hardware complexity of modern machines makes the design of adequate programming models crucial for jointly ensuring performance, portability, and productivity in high-performance computing (HPC). Sequential task-based programming models paired with advanced runtime systems allow the programmer to write a sequential algorithm independently of the hardware architecture in a productive and portable manner, and let a third party software layer -the runtime system- deal with the burden of scheduling a correct, parallel execution of that algorithm to ensure performance. Many HPC algorithms have successfully been implemented following this paradigm, as a testimony of its effectiveness. Developing algorithms that specifically require fine-grained tasks along this model is still considered prohibitive, however, due to per-task management overhead [1], forcing the programmer to resort to a less abstract, and hence more complex "task+X" model. We thus investigate the possibility to offer a tailored execution model, trading dynamic mapping for efficiency by using a decentralized, conservative in-order execution of the task flow, while preserving the benefits of relying on the sequential taskbased programming model. We propose a formal specification of the execution model as well as a prototype implementation, which we assess on a shared-memory multicore architecture with several synthetic workloads. the results show that under the condition of a proper task mapping supplied by the programmer, the pressure on the runtime system is significantly reduced and the execution of fine-grained task flows is much more efficient.

关键词： Runtime Computational modeling Software algorithms computer architecture Programming Hardware Software

来源：评论

学校读者我要写书评

暂无评论

Comparison of Backbones for Semantic Segmentation of X-Ray Microtomography Images using Convolutional Deep Neural architecture 10

Comparison of Backbones for Semantic Segmentation of X-Ray M...

引用

17th INDIACom;2023 10th International Conference on computing for Sustainable Global Development, INDIACom 2023

作者： Sahithya, Gandu Harshitha, K. Priya, Bhuvanagiri V Hari Sirisha, B. Engineering College Department of Computer Science and Engineering India Engineering College Department of Information Technology India

ISBN: (纸本)9789380544472

Mineral processing domains are rapidly expanding and gaining popularity as a result of their ability to achievesubmicron resolution with high contrast on the majority of mineral stages within mining ores. the tomographic (X-Ray microscopy (XRM)) methodology enables very rich statistics on grain sizes and provides fundamental mineralogy differentiation. It is observed that delicate variations in mineral composition result in grayscale reconstructed images that are so close to each other that they are indistinct to conventional segmentation methods. An X-Ray microscopy (XRM) image of sandstone typically consists of minerals with very intricate microstructures, which makes conventional semantic segmentation algorithms inefficient. the authors compare and analyse the weightage of various classification deep networks as the backbone of the semanticsegmentation network (SSN) in this study. the performance of the VGG-16, Resnet-34, and Inception-v3 nets as the Unet and Linknet backbones were examined using a CVPR Deep Globe sandstone dataset. When VGG-16 is employed as the backbone of SSN, it performs better at extracting high-intensity pyrites from the image. When Inception-v3 is used as the backbone of SSN, it can extract even small and average intensity variations, such as clay. Using Resnet-50 as a backbone, all minerals with very complex microstructures can be extracted very efficiently. © 2023 Bharati Vidyapeeth, New Delhi.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Federated Learning Aggregation that Integrates Clustering and Momentum Correction Based on Historical high-Quality Gradients 21

A Federated Learning Aggregation that Integrates Clustering ...

引用

21st IEEE International symposium on Parallel and Distributed Processing with Applications, 13th IEEE International Conference on Big Data and Cloud computing, 16th IEEE International Conference on Social computing and Networking and 13th International Conference on Sustainable computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Xu, Jian Guo, Bing Chen, Fei Shen, Yan Dai, Shengxin Dai, Cheng Sichuan University School of Computer Science Chengdu China Chengdu University of Information Technology School of Computer Science Chengdu China

ISBN: (纸本)9798350329223

Nowadays, federated learning emerges as a new technique to preserve data privacy in distributed machine learning by limiting local data to the client and only aggregating model parameters from multiple parties on the server side. However, it also faces consequent threats from malicious clients, and some methods have been proposed without sufficient defense capability and efficient aggregation. To address these problems, we propose a novel aggregation algorithm, which incorporates PCA operation and clustering to select the excellent client in each round;and brings in a momentum-accelerated historical quality gradient-based approach to increase the weight when the performances of the global model are consistently degraded, which can achieve faster convergence on the basis of ensuring that the update direction of the global model is not manipulated by malicious nodes. Experimental results on the MNIST datasets show that our algorithm is only 0.93% lower than FedAvg under a no-attack scenario, and exhibits faster convergence compared to Krum, Multi-Krum, and Trimmed Mean Aggregation, and the accuracy improves 10.68%,35.83%,19.65% under Black-Box Edge Attack, Sign Flipping Attack, and Same Value Attack. © 2023 IEEE.

关键词： Momentum

来源：评论

学校读者我要写书评

暂无评论

VAQEM: A Variational Approach to Quantum Error Mitigation 28

VAQEM: A Variational Approach to Quantum Error Mitigation

引用

28th Annual IEEE International symposium on high-performance computer architecture (HPCA)

作者： Ravi, Gokul Subramanian Smith, Kaitlin N. Gokhale, Pranav Mari, Andrea Earnest, Nathan Javadi-Abhari, Ali Chong, Frederic T. Univ Chicago Chicago IL 60637 USA Supertech Washington DC USA Unitary Fund Washington DC USA IBM TJ Watson Res Ctr IBM Quantum Yorktown Hts NY USA

ISBN: (纸本)9781665420273

Variational Quantum Algorithms (VQA) are one of the most promising candidates for near-term quantum advantage. Traditionally, these algorithms are parameterized by rotational gate angles whose values are tuned over iterative execution on quantum machines. the iterative tuning of these gate angle parameters make VQAs more robust to a quantum machine's noise profile. However, the effect of noise is still a significant detriment to VQA's target estimations on real quantum machines - they are far from ideal. thus, it is imperative to employ effective error mitigation strategies to improve the fidelity of these quantum algorithms on near-term machines. While existing error mitigation techniques built from theory do provide substantial gains, the disconnect between theory and real machine execution characteristics limit the scope of these improvements. thus, it is critical to optimize mitigation techniques to explicitly suit the target application as well as the noise characteristics of the target machine. We propose VAQEM, which dynamically tailors existing error mitigation techniques to the actual, dynamic noisy execution characteristics of VQAs on a target quantum machine. We do so by tuning specific features of these mitigation techniques similar to the traditional rotation angle parameters by targeting improvements towards a specific objective function which represents the VQA problem at hand. In this paper, we target two types of error mitigation techniques which are suited to idle times in quantum circuits: single qubit gate scheduling and the insertion of dynamical decoupling sequences. We gain substantial improvements to VQA objective measurements a mean of over 3x across a variety of VQA applications, run on IBM Quantum machines. More importantly, while we study two specific error mitigation techniques, the proposed variational approach is general and can be extended to many other error mitigation techniques whose specific configurations are hard to select a pri

关键词： quantum computing noisy intermediate-scale quantum variational quantum algorithms variational quantum eigensolver error mitigation dynamic decoupling gate scheduling

来源：评论

学校读者我要写书评

暂无评论

Analyzing HPC Monitoring Data With a View Towards Efficient Resource Utilization

Analyzing HPC Monitoring Data With a View Towards Efficient ...

引用

International symposium on computer architecture and high performance computing (SBAC-PAD)

作者： Samuel Maloney Estela Suarez Norbert Eicker Filipe Guimarães Wolfgang Frings Jülich Supercomputing Centre Forschungszentrum Jülich Germany Institute of Computer Science University of Bonn Germany School of Mathematics and Natural Sciences University of Wuppertal Germany

ISBN: (数字)9798350356168

ISBN: (纸本)9798350356175

Compute nodes in modern HPC systems are growing in size and their hardware has become ever more diverse. Still, many HPC centers allocate the resources of full nodes exclusively to avoid contention, despite the associated risk of underutilization. this paper describes a thorough resource utilization study of CPU and GPU compute and memory capacity, and interconnect bandwidth on JUWELS, a mature leadership-class modular supercomputer, with the aim of identifying opportunities for improving utilization through advanced scheduling and node sharing. Separate analysis of CPU-only and GPU-accelerated nodes finds that CPU compute usage is already close to optimal for the CPU-only nodes, whereas there is plenty of scope for co-scheduling CPU-based jobs on GPU-accelerated nodes. Memory capacity and node-level interconnect bandwidth are sufficient to provision co-scheduled jobs. We analyze multiple one-month datasets to validate robustness of conclusions over time and compare with previous studies on other systems to establish generalizability of results.

关键词： Runtime high performance computing Graphics processing units Bandwidth computer architecture Supercomputers Robustness Hardware Resource management Monitoring

来源：评论

学校读者我要写书评

暂无评论

OWETC: Open world encrypted traffic classification based on semi-supervised class incremental learning 21

OWETC: Open world encrypted traffic classification based on ...

引用

作者： Zhao, Meng Di, Zhu Xia, Zhe Tian, Jing Xiang, Jianwen Jun, Lin China Electronic Product Reliability and Environmental Testing Research Institute Guangdong China Wuhan University of Technology Hubei Key Laboratory of Transportation of Internet of Things School of Computer Science and Artifical Intelligence Wuhan China Wuhan University of Technology Chongqing Research Institute Chongqing China

ISBN: (纸本)9798350329223

To address the dynamic changes in network environments and effectively tackle the open-world traffic classification challenge, we introduce a novel approach. Conventional traffic classification methods are constrained by their reliance on closed traffic sets and offline training scenarios, limiting their ability to recognize a fixed number of traffic classes and identify unknown traffic types. Our solution involves a traffic classification method founded on a combination of semi-supervised class-incremental learning and open-set recognition. To mitigate the issue of catastrophic forgetting in class-incremental learning, we leverage a substantial amount of external unlabeled traffic data. Specifically, we employ a class anchor clustering loss classifier with a discriminator to select external unlabeled traffic data as replay samples. this approach effectively mitigates the problem of catastrophic forgetting in class-incremental learning. Furthermore, the class anchor clustering loss is employed for incremental model training, enabling the model to identify unknown traffic data. We validate the performance of the model on encrypted traffic classical datasets, and the experimental results show that our method significantly outperforms traffic classification models based on other incremental learning methods. In particular, OWETC has a 3.28% performance improvement in average accuracy over multiple incremental sessions compared to the best baseline. Additionally, our enhanced open-set sampler displays a notable 17% reduction in false alarm rates with only a minor 1% accuracy degradation, surpassing the capabilities of existing open-set recognition methods. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Improving the Energy Efficiency of Real-time DNN Object Detection via Compression, Transfer Learning, and Scale Prediction 16

Improving the Energy Efficiency of Real-time DNN Object Dete...

引用

16th IEEE International Conference on Networking, architecture and Storage (NAS)

作者： Biswas, Debojyoti Rahman, M. M. Mahabubur Zong, Ziliang Tesic, Jelena Texas State Univ Dept Comp Sci San Marcos TX 78666 USA

ISBN: (数字)9781665454087

ISBN: (纸本)9781665454087

In recent years, computational accessibility has enabled the use of Deep Neural Network (DNN) for computer vision applications on devices with limited computational resources. We focus on the real-time object detection algorithms deployed on UAV-friendly devices. the hardware deployed on UAV must be lightweight and thus limited in processing power, memory, and storage capacity. Lightweight modeling architecture does not suffice for high-recall reconnaissance applications. In this paper, we propose to reduce power consumption of YOLOv5 DNN architecture. We decided to use compressed convolutional technique, transfer learning, backbone shrinkage, and scale prediction to reduce the number of learnable parameters from the YOLOvSmodel. Our approach reduced the size of the model significantly and lowered the power consumption in turn. GPU memory and the Billion Floating-Point Operations Per Second (GFLOPS) for the YOLOvSmodel will keep the performance measure of the model as the baseline state-of-the-art. the best resulting model has a 63.86% mean average precision (mAP) and a GFLOPS of 97.7 on "DIOR", an overhead imagery data set. the proposed approach has lowered GPU memory consumption of the model by 34% and lowered the energy consumption by 10 Watts compared to the baseline model.

关键词： Energy Efficiency YOLOv5 real-time object detection Mobile computing UAV

来源：评论

学校读者我要写书评

暂无评论

Optimal Time and Energy-Aware Client Selection Algorithms for Federated Learning on Heterogeneous Resources

Optimal Time and Energy-Aware Client Selection Algorithms fo...

引用

International symposium on computer architecture and high performance computing (SBAC-PAD)

作者： Alan L. Nunes Cristina Boeres Lúcia M. A. Drummond Laércio L. Pilla Fluminense Federal University Niterói Brazil University of Bordeaux CNRS Bordeaux INP Inria LaBRI Talence France

ISBN: (数字)9798350356168

ISBN: (纸本)9798350356175

Federated Learning systems allow training machine learning models distributed across multiple clients, each one using private local data. Iteratively, the clients send their training contributions to a server, which performs a merge to produce an enhanced global model. Due to resource and data heterogeneity, client selection is crucial to optimize the system efficiency and improve the global model generalization. Selecting more clients is likely to increase the overall energy consumption, while a small number of clients may decline the performance of the trained model or require longer training time. We propose two time- and energy-aware client selection algorithms, MEC and ECMTC, which are proven regarding their optimality and evaluated against state-of-the-art algorithms on an extensive series of experiments in both simulation and HPC platform scenarios. the results indicate the benefits of jointly optimizing the time and energy consumption metrics using our proposals.

关键词： Training Measurement Energy consumption Federated learning high performance computing Distributed databases computer architecture Data models Servers Proposals

来源：评论

学校读者我要写书评

暂无评论

Industrial 5G Private Network: architectures, Resource Management, Challenges, and Future Directions 16

Industrial 5G Private Network: Architectures, Resource Manag...

引用

16th IEEE International Conference on Advanced Trends in Radioelectronics, Telecommunications and computer Engineering, TCSET 2022

作者： Ruoxi, Wang Beshley, Halyna Lingyu, Yan Urikova, Oksana Beshley, Mykola Kuzmin, Oleh Wuhan Fiberhome Technical Services Co. Ltd Wuhan China Lviv Polytechnic National University Department of Telecommunications 12 Bandery Str. Lviv Ukraine School of Computer Science Hubei University of Technology Wuhan China Comenius University in Bratislava Faculty of Management Department of Information Systems Bratislava82005 Slovakia Lviv Polytechnic National University Department of Management and International Business 12 Bandery Str. Lviv Ukraine

ISBN: (纸本)9781665468619

Industry 4.0, artificial intelligence, and the Internet of things are fundamental technological advances for the development of smart manufacturing. A properly designed network infrastructure is an important solution for making the manufacturing and logistics processes more flexible, autonomous and efficient. the 5G communications standard opens up important new perspectives for the deployment of industrial private networks. this paper has considered the main problems encountered in traditional networks for deploying smart manufacturing. We have presented a 5G network architecture based on the latest technologies, such as non-orthogonal multiple access, SDN, NFV and cloud computing. the use of these technologies in combination with network slicing provides high reliability and security of the communication channels of industrial networks, as well as stability and isolation from external public networks. For this purpose, we have developed a new method for the dynamic radio resource management between the network slices, taking into account their importance. We developed a method for prioritizing radio resources within a slice, considering the importance of industrial services. We proved that the proposed methods provide ultra-low latency and high reliability for priority messages in the industrial 5G network. © 2022 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

An Error-Resilient Compute-in-Memory 3D FPCA architecture for high-performance Floating-Point Operations

An Error-Resilient Compute-in-Memory 3D FPCA Architecture fo...

引用

IEEE International symposium on Quality Electronic Design

作者： Hasita Veluri Dilip Vasudevan Lawrence Berkeley National Laboratory Applied Mathematics and Computational Research Berkeley CA USA

ISBN: (数字)9798331509422

ISBN: (纸本)9798331509439

We present a runtime reconfigurable Field Programmable Crossbar Array (FPCA) architecture designed to enable high-accuracy multiplication and division alongside high through-put ML operations. Our innovative in-memory compute-based multiplier and divider designs employing novel partial product reduction techniques support expandable multi-precision floating-point operations. Leveraging advanced interface and peripheral design techniques, our error-resilient FPCA achieves 0% error for in-memory digital operations, demonstrating remarkable robustness and reliability despite device and crossbar irregularities. Additionally, our architecture enhances neural network training throughput by 63.8 × and improves power efficiency by 5.18 ×compared to state-of-the-art memristor-based accelerators. this work paves the way for future developments in high-performance, error-resistant in-memory computing solutions.

关键词： Training three-dimensional displays Runtime computer architecture Artificial neural networks In-memory computing throughput Reliability engineering Robustness System-on-chip

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共444页 << < 21 22 23 24 25 26 27 28 29 30 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：