检索结果-内蒙古大学图书馆

arXiv 2017年

作者： Yong-Xian, Wang Li-Lun, Zhang Yong-Gang, Che Chuan-Fu, Xu Wei, Liu Hua-Yong, Liu Zheng-Hua, Wang College of Computer National University of Defense Technology Changsha410073 China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha410073 China State Key Laboratory of Aerodynamics Mianyang621000 China

In this paper, an improved algorithm is proposed for the reconstruction of singularity connectivity from the available pairwise connections during preprocessing phase. To evaluate the performance of our algorithm, an in-house CFD code, in which high-order finite-difference method for spatial discretization, running on the Tianhe-1A supercomputer is employed. Test cases with a varied amount of mesh points are chosen, and the test results indicate that the improved singular connection reconstruction algorithm can achieve a 2000× speedup at least compared with the naive search method adopt in the former version of our code. Moreover, the parallel efficiency can be benefited from the strategy of local communication based on the new algorithm. Copyright © 2017, The Authors. All rights reserved.

关键词： Finite difference method

来源：评论

学校读者我要写书评

暂无评论

Collaborative deep learning across multiple data centers

arXiv

引用

arXiv 2018年

作者： Xu, Kele Mi, Haibo Feng, Dawei Wang, Huaimin Chen, Chuan Zheng, Zibin Lan, Xu National Key Laboratory of Parallel and Distributed Processing Changsha China College of Computer National University of Defense Technology Changsha China School of Data and Computer Science Sun Yat-Sen University Guangzhou China Queen Mary University of London London United Kingdom

Valuable training data is often owned by independent organizations and located in multiple data centers. Most deep learning approaches require to centralize the multi-datacenter data for performance purpose. In practice, however, it is often infeasible to transfer all data to a centralized data center due to not only bandwidth limitation but also the constraints of privacy regulations. Model averaging is a conventional choice for data parallelized training, but its ineffectiveness is claimed by previous studies as deep neural networks are often non-convex. In this paper, we argue that model averaging can be effective in the decentralized environment by using two strategies, namely, the cyclical learning rate and the increased number of epochs for local model training. With the two strategies, we show that model averaging can provide competitive performance in the decentralized mode compared to the data-centralized one. In a practical environment with multiple data centers, we conduct extensive experiments using state-of-the-art deep network architectures on different types of data. Results demonstrate the effectiveness and robustness of the proposed method. Copyright © 2018, The Authors. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Building a lightweight container-based experimental platform for hpc education 10th

Building a lightweight container-based experimental platform...

引用

10th International Symposium on parallel Architectures, Algorithms and Programming, PAAP 2019

作者： Wang, Zelong Wu, Di Luo, Zhenxiao Du, Yunfei School of Data and Computer Science Sun Yat-sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China National Supercomputer Center in Guangzhou Guangzhou China

HPC (High Performance Computing) is of great significance due to its excellent performance in computing acceleration. However, unlike other techniques in computer science, learning HPC requires advanced computing resources such as large-scale clusters which directly increase the cost of study for students. To help students to learn HPC programming easily, we design and develop a lightweight container-based experimental platform to provide students with easily accessible and customizable HPC practice environments. In our platform, we integrate multiple practical functional modules for students, teachers, and administrators respectively. It is convenient for a user to access high-performance computing resources via a web portal, use highly customizable basic environments and have nice graphical hands-on interactive HPC learning experiences from our platform. © Springer Nature Singapore Pte Ltd. 2020.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Accelerating MatLab code using GPU: A review of tools and strategies

Accelerating MatLab code using GPU: A review of tools and st...

引用

2011 2nd International Conference on Artificial Intelligence, Management science and Electronic Commerce, AIMSEC 2011

作者： Zhang, Baida Xu, Shuai Zhang, Feng Bi, Yuan Huang, Linqi National Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China Department of Information Engineering Academy of Armored Force Engineering Beijing 100000 China Electronic Information Industry Co. Ltd Beijing 100085 China Naval Academy of Armament Beijing 100036 China School of Resources and Safety Engineering Central South University Changsha 410083 China

ISBN: (纸本)9781457705366

Lots of toolboxes of accelerating MatLab using GPU are available now[1], but, users are confused by which toolbox is best suitable for a particular task. Three toolboxes-Jacket, GPUmat, and parallel Computing Toolbox of MatLab are selected. For each toolbox, its advantages and pitfalls are reviewed, with an aim to allow the reader to identify which toolbox is appropriate for a given task. Strategies of whether a function should execute on GPU are given after a formula analysis. The analysis is also a framework for program automatically decides which function is cost-efficient to execute on GPU. A series of benchmark of different types of computing, including data transfer between GPU and CPU, data matrix Generation, matrix operation and GPU functions were tested in all three toolboxes. And the results show that Jacket is the best one. Some advices to improve the performance of toolboxes are given in the end. © 2011 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Efficient Approach for Repeated Patterns Mining Based on Indent Shape of HTML Documents

Efficient Approach for Repeated Patterns Mining Based on Ind...

引用

International Conference on Cyber-Enabled distributed Computing and Knowledge Discovery, CyberC

作者： Yanxu Zhu Gang Yin Huaimin Wang Dianxi Shi Xiang Rao Lin Yuan School of computer National Laboratory of Parallel and Distributed Processing National University of Defense Technology Changsha Hunan China College of Electronic Technology Information Engineering University Zhengzhou Henan China

Mining of repeated patterns from HTML documents is the key step towards Web-based data mining and knowledge extraction. Many web crawling applications need efficient repeated patterns mining techniques to generate their wrapper automatically. Existing approaches such as tree matching and string matching can detect repeated patterns with high precision, but their performance is still a challenge for practical web crawling applications. In this paper, we propose an efficient approach for mining repeated patterns based on indent shape of HTML document. Indent shape is a novel and simple model of HTML document, in which tandem repeated waves have strong association with the repeated patterns to be detected. By scanning an indent shape with a horizontal indent-line from bottom to top, the tandem repeated waves are identified by filtering the wave segments with low self-similarities. After that the boundary of HTML code corresponding to repeated patterns can be identified, which could be transformed to regular expressions formal-defined easily. Extensive experiments on two practical data sets retrieved from Internet show that our approach achieves high efficiency significantly, and its precision performance is also generally better than the existing approaches.

关键词： Shape HTML Data mining Web pages Vectors Visualization Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

Sparse Matrix Reordering Method Selection with parallel Computing and Deep Learning

Sparse Matrix Reordering Method Selection with Parallel Comp...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Rui Xia Jihu Guo Huajian Zhang Shun Yang Qinglin Wang Jie Liu College of Computer Science and Techonology National University of Defense Technology Changsha China Laboratory of Digitizing Software for Frontier Equipment National University of Defense Technology Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Sparse matrix reordering is an important step in Cholesky decomposition. By reordering the rows and columns of the matrix, the time of computation and storage cost can be greatly reduced. With the proposal of various reordering algorithms, the selection of suitable reordering methods for various matrices has become an important research topic. In this paper, we propose a method to predict the optimal reordering method by visualizing sparse matrices in chunks in a parallel manner and feeding them into a deep convolutional neural network. The results show that the theoretical performance can reach 95% of the optimal performance, the prediction accuracy of the method can reach up to 85%, the parallel framework achieves an average speedup ratio of 11.35 times over the serial framework, and the performance is greatly improved compared with the traversal selection method on large sparse matrices.

关键词： Deep learning Visualization Accuracy Costs Neural networks parallel processing Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

Deep Isolation Forest for Anomaly Detection

arXiv

引用

arXiv 2022年

作者： Xu, Hongzuo Pang, Guansong Wang, Yijie Wang, Yongjun The College of Computer National University of Defense Technology Changsha410073 China The Science and Technology on Parallel and Distributed Processing Laboratory China The School of Computing and Information Systems Singapore Management University 178902 Singapore

Isolation forest (iForest) has been emerging as arguably the most popular anomaly detector in recent years due to its general effectiveness across different benchmarks and strong scalability. Nevertheless, its linear axis-parallel isolation method often leads to (i) failure in detecting hard anomalies that are difficult to isolate in high-dimensional/non-linear-separable data space, and (ii) notorious algorithmic bias that assigns unexpectedly lower anomaly scores to artefact regions. These issues contribute to high false negative errors. Several iForest extensions are introduced, but they essentially still employ shallow, linear data partition, restricting their power in isolating true anomalies. Therefore, this paper proposes deep isolation forest. We introduce a new representation scheme that utilises casually initialised neural networks to map original data into random representation ensembles, where random axis-parallel cuts are subsequently applied to perform the data partition. This representation scheme facilitates high freedom of the partition in the original data space (equivalent to non-linear partition on subspaces of varying sizes), encouraging a unique synergy between random representations and random partition-based isolation. Extensive experiments show that our model achieves significant improvement over state-of-the-art isolation-based methods and deep detectors on tabular, graph and time series datasets;our model also inherits desired scalability from iForest. © 2022, CC BY.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Optimistic data consistency maintenance method based on data dependence

引用

Jisuanji Xuebao/Chinese Journal of computers 2008年第5期31卷 741-753页

作者： Zhou, Jing Wang, Yi-Jie Li, Si-Kun National Key Laboratory for Parallel and Distributed Processing School of Computer National University of Defense Technology Changsha 410073 China

This paper proposes an optimistic data consistency method according to the question about data dependence in data consistency. In the method, data object is partitioned into data blocks by fixed size as the basic unit of data management. Updates are compressed by Bloom filter technique and propagated in double-path. Negotiation algorithms detect and reconcile update conflicts, and dynamic data management algorithms accommodate dynamic data processing. The results of the performance evaluation show that it is an efficient method to achieve consistency, good dynamic property, and strong robustness when choosing the size of data block appropriately. At the same time, a feasible way is put forward on how to choose appropriate data block size.

关键词： Data processing

来源：评论

学校读者我要写书评

暂无评论

Improving unsupervised domain adaptation by reducing bi-level feature redundancy

arXiv

引用

arXiv 2020年

作者： Wang, Mengzhu Zhang, Xiang Lan, Long Wang, Wei Tan, Huibin Luo, Zhigang Science and Technology on Parallel and Distributed Laboratory College of Computer National University of Defense Technology Changsha China Institute for Quantum State Key Laboratory of High Performance Computing National University of Defense Technology Changsha China DUT-RU International School of Information Science & Engineering Dalian University of Technology DalianLiaoning116000 China Department of Science and Technology on Parallel and Distributed Processing National University of Defense Technology Changsha China

Reducing feature redundancy has shown beneficial effects for improving the accuracy of deep learning models, thus it is also indispensable for the models of unsupervised domain adaptation (UDA). Nevertheless, most recent efforts in the field of UDA ignore this point. Moreover, main schemes realizing this in general independent of UDA purely involve a single domain, thus might not be effective for cross-domain tasks. In this paper, we emphasize the significance of reducing feature redundancy for improving UDA in a bi-level way. For the first level, we try to ensure compact domain-specific features with a transferable decorrelated normalization module, which preserves specific domain information whilst easing the side effect of feature redundancy on the sequel domain-invariance. In the second level, domain-invariant feature redundancy caused by domain-shared representation is further mitigated via an alternative brand orthogonality for better generalization. These two novel aspects can be easily plugged into any BN-based backbone neural networks. Specifically, simply applying them to ResNet50 has achieved competitive performance to the state-of-the-arts on five popular benchmarks. Our code will be available at https://***/dreamkily/gUDA. © 2020, CC BY.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A mesh quality discrimination method based on convolutional neural network

A mesh quality discrimination method based on convolutional ...

引用

IEEE International Conference on Artificial Intelligence and computer Applications (ICAICA)

作者： Zhenpeng Xu Xinhai Chen Lihua Chi Jie Liu Chunye Gong Science and Technology on Parallel and Distributed Processing Laboratory & Laboratory of Software Engineering for Complex Systems National University of Defense Technology Changsha China College of Computer Science and Electronic Engineering Hunan University Changsha China

ISBN: (数字)9781728170053

ISBN: (纸本)9781728170060

With the rapid development of highperformance computing, computational fluid dynamics (CFD) has become an important part of hydrodynamics and aerodynamics. Mesh quality is the key factor that affects the accuracy and efficiency of CFD numerical calculation. However, the current the process of mesh quality discrimination is very time-consuming. The manpower time needed for this process takes up a large proportion in the whole numerical calculation process. A large number of artificial intelligence algorithms have been put forward to replace the human to efficiently complete all kinds of tedious tasks. In this paper, we propose a convolutional neural network (CNN) based mesh quality discrimination method, MeshNet. MeshNet uses residual neural network structure to learn mesh features and automatically judge the mesh quality. The experimental results show that the proposed network can greatly save labor time cost and achieve an accuracy of 94.41% for mesh quality discrimination.

关键词： Convolution Computational fluid dynamics Convolutional neural networks Jacobian matrices Artificial intelligence distributed processing Software engineering

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：