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

10,083 篇 会议
269 篇 期刊文献
38 册 图书

馆藏范围

10,390 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

7,285 篇 工学
- 5,755 篇 计算机科学与技术...
- 3,640 篇 软件工程
- 1,366 篇 信息与通信工程
- 1,287 篇 电气工程
- 877 篇 控制科学与工程
- 825 篇 电子科学与技术（可...
- 465 篇 机械工程
- 462 篇 生物工程
- 392 篇 光学工程
- 370 篇 动力工程及工程热...
- 365 篇 生物医学工程（可授...
- 288 篇 仪器科学与技术
- 258 篇 化学工程与技术
- 239 篇 交通运输工程
- 230 篇 网络空间安全
- 206 篇 材料科学与工程（可...
- 193 篇 土木工程
- 188 篇 建筑学
- 184 篇 安全科学与工程
2,754 篇 理学
- 1,410 篇 数学
- 868 篇 物理学
- 503 篇 生物学
- 362 篇 统计学（可授理学、...
- 342 篇 化学
- 240 篇 系统科学
1,333 篇 管理学
- 819 篇 管理科学与工程(可...
- 629 篇 图书情报与档案管...
- 366 篇 工商管理
444 篇 医学
- 346 篇 临床医学
- 192 篇 基础医学(可授医学...
186 篇 教育学
143 篇 法学
126 篇 经济学
48 篇 农学
41 篇 军事学
31 篇 文学
11 篇 艺术学
1 篇 哲学
1 篇 历史学

主题

1,028 篇 program processo...
729 篇 computer archite...
418 篇 computational mo...
390 篇 hardware
343 篇 high performance...
258 篇 parallel process...
237 篇 computer science
193 篇 optimization
172 篇 benchmark testin...
164 篇 concurrent compu...
163 篇 application soft...
159 篇 runtime
156 篇 computer vision
149 篇 bandwidth
132 篇 deep learning
129 篇 registers
129 篇 libraries
122 篇 scalability
122 篇 parallel program...
121 篇 throughput

机构

70 篇 college of compu...
57 篇 school of comput...
56 篇 university of ch...
45 篇 school of comput...
34 篇 department of co...
32 篇 college of compu...
31 篇 univ chinese aca...
30 篇 school of comput...
30 篇 tsinghua univers...
27 篇 school of comput...
25 篇 school of data a...
23 篇 college of compu...
23 篇 computer network...
23 篇 high energy astr...
22 篇 graduate school ...
21 篇 graduate school ...
20 篇 center for astro...
20 篇 department of ph...
20 篇 department of ph...
20 篇 the graduate sch...

作者

25 篇 tokuno h.
24 篇 ikeda d.
23 篇 belz j.w.
22 篇 jui c.c.h.
22 篇 oshima a.
22 篇 tsuda r.
22 篇 kim j.h.
22 篇 bergman d.r.
22 篇 allen m.
22 篇 matsuyama t.
22 篇 wong t.
22 篇 ishii t.
22 篇 kim s.w.
22 篇 takamura m.
22 篇 yashiro k.
22 篇 kadota k.
22 篇 mayta r.
22 篇 ozawa s.
22 篇 kasami s.
22 篇 rodriguez d.c.

语言

9,499 篇 英文
851 篇 其他
67 篇 中文
3 篇 西班牙文
2 篇 土耳其文

检索条件"任意字段=Conference on High Performance Computer Modernization Program"

共 10390 条记录，以下是4141-4150 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：

Deep Hierarchical Multi-label Classification of Chest X-ray Images 2

Deep Hierarchical Multi-label Classification of Chest X-ray ...

引用

2nd International conference on Medical Imaging with Deep Learning (MIDL)

作者： Chen, Haomin Miao, Shun Xu, Daguang Hager, Gregory D. Harrison, Adam P. NVIDIA AI Infra Bethesda MD 20814 USA Johns Hopkins Univ Dept Comp Sci Baltimore MD 21218 USA

Chest X-rays (CXRs) are a crucial and extraordinarily common diagnostic tool, leading to heavy research for computer-Aided Diagnosis (CAD) solutions. However, both high classification accuracy and meaningful model predictions that respect and incorporate clinical taxonomies are crucial for CAD usability. To this end, we present a deep Hierarchical Multi-Label Classification (HMLC) approach for CXR CAD. Different than other hierarchical systems, we show that first training the network to model conditional probability directly and then refining it with unconditional probabilities is key in boosting performance. In addition, we also formulate a numerically stable cross-entropy loss function for unconditional probabilities that provides concrete performance improvements. To the best of our knowledge, we are the first to apply HMLC to medical imaging CAD. We extensively evaluate our approach on detecting 14 abnormality labels from the PLCO dataset, which comprises 198;000 manually annotated CXRs. We report a mean Area Under the Curve (AUC) of 0:887, the highest yet reported for this dataset. These performance improvements, combined with the inherent usefulness of taxonomic predictions, indicate that our approach represents a useful step forward for CXR CAD.

关键词： hierarchical multi-label classification chest x-ray computer aided diagnosis

来源：评论

学校读者我要写书评

暂无评论

COSMOS - Framework for Combining Optimization and Simulation Software 22

COSMOS - Framework for Combining Optimization and Simulation...

引用

22nd International conference on Control Systems and computer Science (CSCS)

作者： Samoila, Cosmin-Gabriel Slusanschi, Emil-Ioan Univ Politehn Bucuresti Dept Comp Sci & Engn Bucharest Romania

ISBN: (纸本)9781728123318

A simulation software is a mathematical model translated into code and it describes a physical system. Given the fact that designing and writing simulation software requires a lot of knowledge about the system we want to model, we will focus on integrating optimization software with existing simulation programs for high energy particles and molecular dynamics. An optimization software is a special kind of software that targets applications and tunes their parameters and input accordingly to the desired result. By performing an optimization step, we are aiming either a lower run time or a better simulation precision. COSMOS is a framework that aims to provide three main functions: to optimize simulation output with respect to program input and parameters, to provide improved scheduling policies for all tasks of the simulation packages on HPC resources, and to offer enhanced sensitivity analysis of application parameters.

关键词： high performance computing optimization simulation software job scheduling

来源：评论

学校读者我要写书评

暂无评论

A Balanced Cost Fault-Tolerant Scheduling Algorithm in Heterogeneous Real-Time Systems

A Balanced Cost Fault-Tolerant Scheduling Algorithm in Heter...

引用

IEEE International conference on high performance Computing and Communications (HPCC)

作者： Jing Liu Yifu Zhang Yang Liu Pei Yang Cen Chen Ziqi Zhu Zeng Zeng College of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan China Institute for Infocomm Research Singapore

ISBN: (纸本)9781728176499;9781728176505

high reliability is required for many safety-critical and real-time applications. It is difficult to avoid permanent faults during the executing process of tasks. Once such faults are not processed in time, real-time tasks cannot be completed within the specified time, which may lead to catastrophic consequences. In order to deal with the problem of permanent fault tolerance in heterogeneous real-time systems, in this paper, we propose a balanced cost fault-tolerant scheduling algorithm. First, the algorithm uses the task computation time and the average communication time of its successor nodes as the ranking factors when calculating task priority, so that the priority of the task with larger average communication time improved; second, a task cost table is established, and tasks are allocated to appropriate processors according to the values in the task cost table in processor allocation stage. Experimental results show that the proposed algorithm can effectively improve the system reliability and reduce the completion time.

关键词： Fault tolerance program processors Scheduling algorithms high performance computing conferences Fault tolerant systems Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Processor Pipelining Method for Efficient Deep Neural Network Inference on Embedded Devices

Processor Pipelining Method for Efficient Deep Neural Networ...

引用

International conference on high performance Computing

作者： Akshay Parashar Arun Abraham Deepak Chaudhary Vikram Nelvoy Rajendiran On-Device AI Research &Development Samsung R&D Institute Bangalore Bengaluru India

Myriad applications of Deep Neural Networks (DNN) and the race for better accuracy have paved the way for the development of more computationally intensive network architectures. Execution of these heavy networks on embedded devices needs highly efficient real-time DNN inference frameworks. But the sequential architecture of popular DNNs makes it difficult to parallelize its operations among different processors. We propose a novel pipelining method pluggable on top of conventional inference frameworks and capable of parallelizing DNN inference on heterogeneous processors without impacting the accuracy. We partition the network into subnets, by estimating the optimal split points, and pipeline these subnets across multiple processors. The results shows that the proposed method achieves up to 68% improvement in the frames per second (FPS) rate of popular network architectures like VGG19, DenseNet-121 and ResNet-152. Moreover, we show that our method can be used to extract even more performance out of high performance chipsets, by better utilizing the capabilities of its AI processor ecosystem. We also showcase that our method can be easily extended to other low performance chipsets, where this additional performance gain is crucial to deploy real-time AI applications. Our results show performance improvement of up to 47% in the FPS rate on these chipsets without the need of specialized AI hardware.

关键词： performance evaluation program processors Ecosystems Neural networks computer architecture Network architecture Real-time systems

来源：评论

学校读者我要写书评

暂无评论

SwFLOW: A dataflow deep learning framework on sunway taihulight supercomputer 21

SwFLOW: A dataflow deep learning framework on sunway taihuli...

引用

21st IEEE International conference on high performance Computing and Communications, 17th IEEE International conference on Smart City and 5th IEEE International conference on Data Science and Systems, HPCC/SmartCity/DSS 2019

作者： Lin, Han Lin, Zeng Diaz, Jose Monsalve Li, Mingfan An, Hong Gao, Guang R. School of Computer Science and Technology University of Science and Technology of China Hefei China University of Delaware Delaware United States

ISBN: (纸本)9781728120584

Deep learning technology is widely used in many modern fields and a number of deep learning models and software frameworks have been proposed. However, it is still very difficult to process deep learning tasks efficiently on traditional high performance computing (HPC) systems with specialized architectures such as Sunway TaihuLight. In this paper, we propose swFLOW: a TensorFlow-based dataflow deep learning framework on Sunway TaihuLight. Based on the performance analysis results on convolutional neural network (CNN), we optimize the convolution layer, reduce the data layout transpose operation and get 10.42x speedup compared to single management processing element (MPE) version. As for distributed training, we use elastic averaging stochastic gradient descent (EASGD) algorithm to reduce communication and use data prefetch to avoid data fetch being a performance bottleneck. On 512 processes, we get a parallel efficiency of 81.01% with communication period τ = 8. Limited by the maximal executable batch size, the current performance of swFLOW is far from optimal. It is very necessary to further optimize using technology like remote direct memory access (RDMA) and model parallelism. © 2019 IEEE.

关键词： Supercomputers

来源：评论

学校读者我要写书评

暂无评论

Compressing encrypted images by using 2D compressed sensing 21

Compressing encrypted images by using 2D compressed sensing

引用

作者： Zhang, Bo Xiao, Di Zhang, Zhenyu Yang, Lei Communication NCO Academy Army Engineering University Chongqing China College of Computer Science Chongqing University Chongqing China

ISBN: (纸本)9781728120584

Compressing encrypted images by using compressed sensing (CS) has received more and more interests in recent years. However, the existing CS-based coding schemes for encrypted images cannot achieve the low computational complexity and high security at the same time. To address this problem, a novel image encryption-then-compression (ETC) system by using 2D CS (2DCS) is proposed in this paper. Firstly, for high security purpose, the entire image is encrypted by using global random permutation. Secondly, for low computational complexity purpose, the encrypted image is compressed by using 2DCS. Lastly, a 2D projected Landweber (2DPL) algorithm is proposed for image reconstruction. Compared with the state-of-the-art CS-based ETC methods, it has two advantages. Firstly, it achieves the low computational complexity and high security at the same time. Secondly, it obtains a significant gain of peak signal-to-noise ratio (PSNR) of the reconstructed-images. © 2019 IEEE.

关键词： Compressed sensing

来源：评论

学校读者我要写书评

暂无评论

Implementing the matrix multiplication with dfc on kunlun small scale computer 20

Implementing the matrix multiplication with dfc on kunlun sm...

引用

20th International conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019

作者： Du, Zheng Zhang, Jing Sha, Shihao Luo, Qiuming Department of Computer Science and Technology Tsinghua University China NHPCC/Guangdong Key Laboratory of Popular HPC Guangdong Province Engineering Center of China-made High Performance Data Computing System College of Computer Science Software Engineering Shenzhen University Shenzhen China

ISBN: (纸本)9781728126166

In this paper, we demonstrate a new dataflow platform of DFC, which can handle the successive dataflow computing passes with tagged data. By implementing the matrix multiplication in DFC, we show that DFC can exploit the parallelism automatically with a much simple dataflow graph constructed by DF functions of DFC. Different from the other dataflow execution platform, DFC support multiple worker threads for one dataflow node of DF functions. By running the matrix multiplication program of DFC on Kunlun system, it was verified that DFC get a reasonable speedup for large scale computing for thread number up to 512. © 2019 IEEE.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Influence of Pier Settlement on Dynamic performance of high-Speed Maglev Train

Influence of Pier Settlement on Dynamic Performance of High-...

引用

Information Science, computer Technology and Transportation (ISCTT), International conference on

作者： Xiaobo Hong Jun Wu Yunzhou Zhang Yongxiang He College of Intelligent Science National University of Defense Technology Changsha China

ISBN: (数字)9781728185750

ISBN: (纸本)9781728185767

The pier settlement leads to the irregularity of highspeed maglev track, which affects the safety and stability of train operation. It is very important to set the safety threshold of pier settlement reasonably and accurately. In this paper, the dynamic model of high-speed maglev train/track interaction system is established. The total response of track beam is divided into static response caused by pier settlement and dynamic response caused by vibration of elastically beam, so a track beam model composed of dynamic displacement model and static displacement model is established. The dynamic performance of the train/track coupling system is evaluated under the conditions of different settlement values and number of settlement piers, and the safety threshold of pier settlement is analyzed from the perspective of passenger comfort and running stability of the train. The results show that the safety threshold of pier settlement is about 16mm for the maglev train with a speed of 400km/h.

关键词： Vibrations Magnetic levitation vehicles Air gaps Transportation Stability analysis Safety Acceleration

来源：评论

学校读者我要写书评

暂无评论

Optimization of jacobi iteration on the Intel Xeon Phi 19

Optimization of jacobi iteration on the Intel Xeon Phi

引用

3rd high performance Computing and Cluster Technologies conference, HPCCT 2019 and the 2nd International conference on Big Data and Artificial Intelligence, BDAI 2019

作者： Yang, Wenxiang Zou, Jiming Deng, Liang China Aerodynamics Research and Development Center Mianyang621000 China School of Computer Science National University of Defense Technology Changsha410073 China

ISBN: (纸本)9781450371858

Jacobi iteration based on finite difference and finite element discrete scheme is a kind of typical stencil computation in scientific computing. In this paper, we analyze the parallel optimization of Jacobi iteration in the real CFD codes on the Intel Many Integrated Core architecture, and get high performance. We use loop fusion, data structure transformation, subroutine and loop unrolling, cache blocking and some other optimization techniques in our implementation. We also collect hardware performance indicators through the open source performance analysis tools, in order to guide and verify the performance optimization on the many-core architectures. Experimental results on Intel Xeon Phi working in the native execution mode show that our Jacobi iteration can achieve 83.47% parallel efficiency and 4.73 speed ratio of vectorization with a 128 ×128 ×256 grid. © 2019 Association for Computing Machinery.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Prediction Methods for Credit Default on Peer to Peer Lending using Machine Learning 4

A Comparison of Prediction Methods for Credit Default on Pee...

引用

4th International conference on computer Science and Computational Intelligence (ICCSCI) - Enabling Collaboration to Escalate Impact of Research Results for Society

作者： Setiawan, Netty Suharjito Diana Bina Nusantara Univ Comp Sci Dept Binus Grad Program Master Comp Sci Jakarta 11480 Indonesia Bina Nusantara Univ Sch Comp Sci Comp Sci Dept Jakarta 11480 Indonesia

Social lending or peer to peer lending (p2p lending) has emerged as a viable digital platform where lenders and borrowers can do business without the involvement of financial institutions. P2p lending has gained significant momentum recently, with some platform has reached billion-dollar loan circulation. However, p2p lending platforms are not free from any form of risks. A higher return on investment for investor comes with a risk of the loan and interest not being repaid. For this purpose, this research proposes a tree-based classification method for predicting whether a loan will go bad or default before the loan is approved. The high dimensionality of the dataset needs to be processed and chosen carefully. This paper proposes a Binary PSO with SVM to perform feature selection for the dataset and Extremely Randomized Tree (ERT) and Random Forest (RF) as the classifiers. In this research, BPSOSVM-ERT and BPSOSVM-RF are compared with several performance metrics. The experimental results show BPSOSVM can produce subset of features without decreasing the performance from the original features and ERT can outperform RF in several performance metrics. (C) 2019 The Authors. Published by Elsevier B.V.

关键词： P2P lending Random forest Extremely randomized tree PSO BPSO

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500页 << < 411 412 413 414 415 416 417 418 419 420 > >>

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：