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

4,545 篇 会议
81 篇 期刊文献
15 册 图书

馆藏范围

4,641 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

2,557 篇 工学
- 2,364 篇 计算机科学与技术...
- 1,157 篇 软件工程
- 494 篇 电气工程
- 412 篇 信息与通信工程
- 217 篇 控制科学与工程
- 212 篇 电子科学与技术（可...
- 104 篇 网络空间安全
- 80 篇 动力工程及工程热...
- 80 篇 生物工程
- 70 篇 机械工程
- 67 篇 光学工程
- 55 篇 生物医学工程（可授...
- 36 篇 建筑学
- 30 篇 化学工程与技术
- 29 篇 环境科学与工程（可...
611 篇 理学
- 372 篇 数学
- 144 篇 物理学
- 93 篇 生物学
- 74 篇 系统科学
- 68 篇 统计学（可授理学、...
- 28 篇 化学
464 篇 管理学
- 346 篇 管理科学与工程(可...
- 133 篇 图书情报与档案管...
- 129 篇 工商管理
- 65 篇 公共管理
110 篇 法学
- 102 篇 社会学
103 篇 医学
- 72 篇 公共卫生与预防医...
- 30 篇 临床医学
65 篇 文学
- 65 篇 新闻传播学
39 篇 经济学
- 38 篇 应用经济学
17 篇 教育学
10 篇 农学
1 篇 艺术学

主题

1,192 篇 computer archite...
425 篇 hardware
344 篇 high performance...
294 篇 computational mo...
255 篇 concurrent compu...
236 篇 application soft...
213 篇 computer science
205 篇 parallel process...
180 篇 distributed comp...
178 篇 costs
171 篇 delay
169 篇 grid computing
168 篇 bandwidth
164 篇 field programmab...
153 篇 resource managem...
153 篇 cloud computing
145 篇 computer network...
142 篇 throughput
129 篇 processor schedu...
125 篇 scalability

机构

22 篇 institute of com...
16 篇 graphic era deem...
14 篇 chitkara univers...
13 篇 university of ch...
11 篇 college of compu...
10 篇 school of comput...
10 篇 computer science...
9 篇 georgia inst tec...
9 篇 computer systems...
9 篇 intel corp santa...
8 篇 department of co...
8 篇 department of el...
8 篇 ibm thomas j. wa...
8 篇 barcelona superc...
8 篇 mathematics and ...
8 篇 quantum universi...
8 篇 中国电子科技集团...
8 篇 carnegie mellon ...
8 篇 mathematics and ...
8 篇 key laboratory o...

作者

14 篇 navaux philippe ...
10 篇 satvik vats
9 篇 vats satvik
9 篇 i. foster
9 篇 d.k. panda
8 篇 dhabaleswar k. p...
8 篇 dongarra jack
8 篇 loh gabriel h.
8 篇 guedes dorgival
8 篇 viktor k. prasan...
8 篇 prasanna viktor ...
8 篇 mutlu onur
7 篇 shiva mehta
7 篇 borin edson
7 篇 chong frederic t...
7 篇 torrellas josep
7 篇 kim nam sung
7 篇 li chao
7 篇 zhou huiyang
7 篇 ferreira renato

语言

4,578 篇 英文
41 篇 其他
19 篇 中文
3 篇 葡萄牙文
1 篇 法文

检索条件"任意字段=15th Symposium on Computer Architecture and High Performance Computing"

共 4641 条记录，以下是4491-4500 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：

Adaptive Fusion-Enhanced CNN architectures for Advanced Image Classification in E-Commerce

Adaptive Fusion-Enhanced CNN Architectures for Advanced Imag...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： Shantanu Kumar Senior Software Engineer Amazon US

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Image classification is a fundamental aspect of modern artificial intelligence, impacting various applications ranging from autonomous vehicles to medical diagnostics. With the development of Convolutional Neural Networks (CNNs), traditional classification methods have been significantly enhanced, improving the accuracy of image analysis by capturing complex patterns within extensive datasets. Despite these advancements, a notable challenge remains in dynamically integrating multiple CNN architectures to fully utilize their collective strengths, especially in adapting efficiently to diverse image characteristics. To address this challenge, this study introduces a novel fusion model incorporating the unique features of three well-known CNNs: VGG16, ResNet50, and MobileNetV2. this model employs a dynamic concatenation strategy that effectively merges each network’s unique capabilities based on the input images’ specific characteristics, thereby improving the model’s adaptability and performance. Designed to leverage the depth and complexity of VGG16, the residual learning capabilities of ResNet50, and the efficiency of MobileNetV2, this robust system excels in accuracy and computational efficiency. Experimental results on the Fashion-MNIST dataset confirm the effectiveness of our fusion model, which achieves a training accuracy of 97.50% and a validation accuracy of 89.27%. this model markedly surpasses traditional single-model approaches, especially in distinguishing more nuanced categories, as shown by detailed evaluations including precision, recall, F1-score, and ROC curve analysis. this research effectively closes a significant gap in adaptive model fusion. It establishes a new benchmark for image classification tasks, presenting substantial implications for future applications that demand high accuracy and efficiency.

关键词： Training Adaptation models Accuracy Computational modeling Scalability computer architecture Convolutional neural networks Vehicle dynamics Image classification Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Vacuum packing: extracting hardware-detected program phases for post-link optimization

Vacuum packing: extracting hardware-detected program phases ...

引用

IEEE/ACM International symposium on Microarchitecture (MICRO)

作者： R.D. Barnes E.M. Nystrom M.C. Merten W.W. Hwu Center for Reliable and High-Performance Computing Department of Electical and Computer Engineering University of Illinois Urbana-Champaign USA

this paper presents Vacuum Packing, a new approach to profile-based program optimization. Instead of using traditional aggregate or summarized execution profile weights, this approach uses a transparent hardware profiler to automatically detect execution phases and record branch profile information for each new phase. the code extraction algorithm then produces code packages that are specially formed for their corresponding phases. the algorithm compensates for the incomplete and often incoherent branch profile information that arises due to the nature of hardware profilers. the technique avoids unnecessary code replication by focusing on hot code, making efficient connections between the original code and the new code, linking code packages at select points to facilitate phase transitions, and providing a platform for efficient optimization. We demonstrate that using a concise set of profile information from a hardware profiler, we can generate code packages, specialized for each phase of execution, that capture more than 80% of the average total program execution. We further show that the approach is very effective in extracting code regions that capture the phasing behavior of programs, that the code size increase is moderate, and that the code regions benefit from sample optimizations.

关键词： Data mining Packaging Reliability engineering Aggregates Phase detection Joining processes computer architecture Java Emulation Libraries

来源：评论

学校读者我要写书评

暂无评论

ASNet: Introducing Approximate Hardware to high-Level Synthesis of Neural Networks

ASNet: Introducing Approximate Hardware to High-Level Synthe...

引用

International symposium on Multiple-Valued Logic

作者： Saman Froehlich Lucas Klemmer Daniel Große Rolf Drechsler DFKI GmbH and Group of Computer Architecture University of Bremen Germany

ISBN: (数字)9781728154060

ISBN: (纸本)9781728154077

Approximate computing is a design paradigm which makes use of error tolerance inherent to many applications in order to trade off accuracy for performance. One classic example for such an application is machine learning with Neural Networks (NNs). Recently, LeFlow, a high-Level Synthesis (HLS) flow for mapping Tensorflow NNs into hardware has been proposed. the main steps of LeFlow are to compile the Tensorflow models into the LLVM Intermediate Representation (IR), perform several transformations and feed the result into a HLS tool. In this work we take HLS-based NN synthesis one step further by integrating hardware approximation. To achieve this goal, we upgrade LeFlow such that (a) the user can specify hardware approximations, and (b) the user can analyze the impact of hardware approximation already at the SW level. Based on the exploration results which satisfy the NN quality expectations, we import the chosen approx. HW components into an extended version of the HLS tool to finally synthesize the NN to Verilog. the experimental evaluation demonstrates the advantages of our proposed ASNet for several NNs. Significant area reductions as well as improvements in operation frequency are achieved.

关键词： Artificial neural networks Tools Hardware design languages Generators Libraries Hardware Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Parallel 2D Delaunay triangulations in HPP and MPI

Parallel 2D Delaunay triangulations in HPP and MPI

引用

International symposium on Parallel and Distributed Processing (IPDPS)

作者： Min-Bin Chen Tyng-Ruey Chuang Jan-Jau Wu Institute of Information Science Academia Sinica Taipei Taiwan

ISBN: (纸本)0769509908

this paper reports on efficient parallel implementations of two-dimensional Delaunay triangulation in high performance Fortran (HPF) and in Message Passing Interface (MPI). Our parallelization algorithm performs subblock triangulation and boundary merge independently at the same time. the sub-block triangulation is by a divide & conquer Delaunay algorithm known for its sequential efficiency, and the boundary triangulation is by an incremental construction algorithm with low overhead. Compared to prior work, our parallelization method is both simple and efficient. In the paper we also describe a solution to the collinear points problem that usually arises in large data sets for divide & conquer algorithm. Our experiences with the HPF implementation show that with careful control of the data distribution, we are able to parallelize the program using HPF's standard directives and extrinsic procedures. Experimental results on several parallel platforms, including an IBM SP2 and a DEC Alpha farm, show that a parallel efficiency of 31%-59% can be achieved for an 8-node distributed memory system. We also compare efficiency of the HPF implementation to that of a similarly hand-coded MPI implementation.

关键词： Parallel algorithms Mesh generation Load management Information science Message passing Concurrent computing Parallel processing Merging

来源：评论

学校读者我要写书评

暂无评论

A Novel Network Selection Mechanism for Vehicle-to-Infrastructure Communication

A Novel Network Selection Mechanism for Vehicle-to-Infrastru...

引用

IEEE International symposium on Dependable, Autonomic and Secure computing (DASC)

作者： Emmanuel Ndashimye Nurul I Sarkar Sayan Kumar Ray School of Engineering Computer and Mathematical Sciences Auckland University of Technology Auckland New Zealand Faculty of Business and Information Technology Manukau Institute of Technology Auckland New Zealand

the commercial deployment of 5G networks require heterogeneous multi-tier, multiple radio access technologies (RATs) to support vehicle-to-infrastructure (V2I) communication with diversified services. Vehicles may need to cross a number of heterogeneous networks of various sizes before reaching the destination. Due to high-speed travel, vehicles may quickly move in and out of the network coverage areas while performing handover. Fast and efficient selection of an appropriate underlying network is critical for seamless handover performance. In this paper we propose a novel network selection mechanism for improved handover performance in V2I communication over heterogeneous wireless network. the idea is for vehicles to self-evaluate a candidate list of access points (AP) that are located in the vehicle movement direction and select the best underlying candidate network based on key criteria, like, the distance between target candidate and the trajectory of the vehicle movement as well as the vehicle mobility information. Fuzzy logic inference system is used to decide whether a target candidate is suitable for handover. Experimental results show that for a vehicle moving at 30km=h, an AP of 100m radius should be located at less than 30m from the road, while this distance is limited to 15m when the vehicle speed is 60km/h.

关键词： Handover Vehicles Delays Fuzzy logic Long Term Evolution Wireless networks

来源：评论

学校读者我要写书评

暂无评论

Workflow Design Analysis for high Resolution Satellite Image Analysis

Workflow Design Analysis for High Resolution Satellite Image...

引用

IEEE International Conference on e-Science and Grid computing

作者： Ioannis Paraskevakos Matteo Turilli Bento Collares Gonçalves Heather J. Lynch Shantenu Jha Department of Electrical and Computer Engineering Rutgers University Piscataway New Jersey Rutgers University Piscataway NJ USA Stony Brook University Stony Brook NY USA Department of Ecology and Evolution Brookhaven National Laboratory Stony Brook NY USA

ISBN: (数字)9781728124513

ISBN: (纸本)9781728124520

Ecological sciences are using imagery from a variety of sources to monitor and survey populations and ecosystems. Very high Resolution (VHR) satellite imagery provide an effective dataset for large scale surveys. Convolutional Neural Networks have successfully been employed to analyze such imagery and detect large animals. As the datasets increase in volume, O(TB), and number of images, O(1k), utilizing high performance computing (HPC) resources becomes necessary. In this paper, we investigate a task-parallel data-driven workflows design to support imagery analysis pipelines with heterogeneous tasks on HPC. We analyze the capabilities of each design when processing a dataset of 3,000 VHR satellite images for a total of 4~TB. We experimentally model the execution time of the tasks of the image processing pipeline. We perform experiments to characterize the resource utilization, total time to completion, and overheads of each design. Based on the model, overhead and utilization analysis, we show which design approach to is best suited in scientific pipelines with similar characteristics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Machine Intelligence at the Crease: A Quintuple-Class CNN-LSTM Model for Umpire Gesture Recognition

Machine Intelligence at the Crease: A Quintuple-Class CNN-LS...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： Satvik Vats Shiva Mehta Computer Science and Engineering Graphic Era Hill University Dehradun India Graphic Era Deemed to be University Dehradun India Centre for Research Impact & Outcome Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

the model is advanced and state-of-the-art. How successful can a methodology be assessed in identifying essential performance practicalities and applying them to all five umpire actions accomplished? the precisions, recall, and F1 scores have been measured correctly; this allows us to quantify the model’s prediction accuracy. For Class 1, which is facile to detect, the precision record was higher than 87.3%, with a recall of $\mathbf{9 7. 3 \%}$ and an F 1 score of 92.1%. A fantastic level of accuracy characterizes class 2, as it reported a high precision of above 87% and a recall of above 94%, with F1-score precision above $\mathbf{9 0 \%}$. the Grade 3 model displays an accuracy of 86.94%, a recall of 84.12%, and an F1-score of 85.50. the precision for Class 4 is 89%, with a sensitivity of 81% and an $F 1$ score of 85%. Its accuracy is 90.43%, and its recall is 87.93%, so the F1 score is also 89.16%. Among the averaging techniques, macro, micro, and weighted averages incorporate class-wise errors and summarise the entire model performance. the macro average indicates that the accuracy, number of articles returned, and F1 score are 88.27 %, 89.02 %, and 88.51%, respectively. the weighted average is concerned with the weighted class distribution, which provides accuracy, recall, and F1 of 88.38%, 88.32%, and 88.23%, respectively. the micro average results from all courses’ total contributions show a global performance of 88.32% in accuracy, recall, and F1 score. Moreover, the accuracy rate of the proposed model will astonish anyone at 88.3%. When we add it up, it is evident that the CNN-LSTM model, which we use to recognize umpire actions automatically, is an accurate statement. this model displays the possibility of a dramatic increase in the accuracy and dependability of the cricket referee.

关键词： Accuracy Sensitivity Neural networks Deep architecture Gesture recognition Predictive models Robustness Convolutional neural networks Machine intelligence Sports

来源：评论

学校读者我要写书评

暂无评论

Accelerating mutual information-based 3D medical image registration with An FPGA computing platform (abstract only) 05

Accelerating mutual information-based 3D medical image regis...

引用

Proceedings of the 2005 ACM/SIGDA 13th international symposium on Field-programmable gate arrays

作者： Jianchun Li Christos Papachristou Raj Shekhar Case Western Reserve University Cleveland OH University of Maryland Baltimore MD

ISBN: (纸本)9781595930293

Mutual information-based 3D image Registration algorithm is shown to be an accurate, robust and more general registration method for medical image processing. However, its potential application in clinic is jeopardized by its high computational costs. Like many other 3D imaging algorithms, 3D registration needs to process vast amount of image data. the intrinsic computing characteristics of the mutual information-based registration algorithm make it even more difficult to manage and cache the input data. In this paper, we introduce our FPGA-based computing platform and detail its application in accelerating the mutual information-based 3D image registration. this platform is designed to accelerate a series of 3D medical imaging algorithms dominated by local operations. It is a System-on-Chip (SoC) architecture with an intelligent data caching system for efficient data fetching and buffering. A reconfigurable multi-pipeline execution unit in the platform is designed to perform high-speed parallel computation. this unit can be customized according to different algorithm requirements to achieve high performance. the simulation results for accelerating the mutual information based registration with our platform show that a speed-up of about 30 can be achieved when compared to a single-CPU computer. Moreover, the platform can be easily reconfigured to accelerate other 3D medical imaging algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Real-Time Wrong-Way Vehicle Detection System with Automatic Number Plate Recognition for Enhanced Road Safety

Real-Time Wrong-Way Vehicle Detection System with Automatic ...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： Krish Gaur Miran Ahmad Siddique Krishna Beernally Nityam Madaan Sandhya Tarwani Artificial Intelligence-Machine Learning Vivekananda Institute of Professional Studies New Delhi India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

In the pursuit of enhancing road safety, the development of real-time detection systems has become imperative. this paper presents a smart approach to wrong-way vehicle detection utilizing computer vision technology for enhanced road safety. the proposed system presents a combination of YOLO for vehicle and license plate detection and OpenCV for license plate recognition, specifically tailored for Indian vehicle number plate dataset. By integrating YOLO’s single-pass object detection capabilities, the system achieves real-time performance, in order to take immediate intervention in traffic violations. Utilizing methods such as centroid tracking and ‘entry-exit’ analysis, the system effectively differentiates between vehicles traveling in correct lanes and those in wrong lanes, enabling immediate enforcement measures. Centroid tracking assigns and tracks centre points of vehicle bounding boxes across frames, ensuring computational efficiency and simplicity in implementation. the ‘entry-exit’ approach utilizes designated lines in video frames to determine the direction of vehicle movement, leading towards accurate wrong-way detection. Moreover, the system leverages EasyOCR for text extraction, ensuring high accuracy in capturing license plate information. this integration of technologies allows for comprehensive monitoring of traffic patterns and swift identification of violations, contributing to proactive enforcement strategies and improved road safety. through extensive experimentation on the Indian vehicle number plate YOLO annotated dataset, the proposed system demonstrates robust performance in real-world scenarios, exhibiting high accuracy and reliability in wrong-way vehicle detection and automatic number plate recognition. this research represents a significant advancement in leveraging computer vision for proactive traffic management and underscores the potential of technology in enhancing road safety measures.

关键词： YOLO computer vision Accuracy Tracking Traffic control Road safety Real-time systems Reliability License plate recognition Monitoring

来源：评论

学校读者我要写书评

暂无评论

Leveraging Advanced Machine Learning Algorithms for Loan Repayment Prediction

Leveraging Advanced Machine Learning Algorithms for Loan Rep...

引用

International Conference on computing and Networking Technology (ICCNT)

作者： Akshat Ranjan Mahendra Kumar Gourisaria Jay Prakash Singh Amiya Ranjan Panda Soumya Ranjan Mishra Varutri Parihar School of Computer Engineering KIIT Deemed to be University Bhubaneshwar Odisha India Dept of Computer Science & Engineering Manipal University Jaipur Jaipur Rajasthan India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

Accurately predicting loan repayment behavior is a critical challenge for financial institutions, which often face high default rates and financial instability due to inaccurate credit assessments. In order to address these problems, this study looks into the application of sophisticated machine learning methods such Random Forests, Decision Trees, Support Vector Machines, and Stacking Classifiers. the models, trained on diverse client variables, significantly outperform traditional credit scoring methods. the findings suggest that integrating these sophisticated models can enhance financial performance, reduce default risks, and enable more informed lending decisions. the study also examines potential avenues for future research, including the application of increasingly sophisticated machine learning techniques and the integration of new data sources to enhance models. this study advances the field of machine learning in finance by providing business and academics with insightful information that could result in stronger and more secure financial systems.

关键词： Support vector machines Machine learning algorithms Soft sensors Computational modeling Stacking Finance Data models Decision trees Random forests Faces

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465页 << < 446 447 448 449 450 451 452 453 454 455 > >>

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：