检索结果-内蒙古大学图书馆

2018 International Joint Conference on Neural Networks, IJCNN 2018

作者： Yu, Hao Zhang, Zhaoning Qin, Zheng Wu, Hao Li, Dongsheng Zhao, Jun Lu, Xicheng Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China College of Electronic and Engineering National University of Defense Technology Changsha China College of Meteorology and Oceanology National University of Defense Technology Changsha China

ISBN: (纸本)9781509060146

Modern object detectors usually suffer from low accuracy issues, as foregrounds always drown in tons of back-grounds and become hard examples during training. Compared with those proposal-based ones, real-time detectors are in far more serious trouble since they renounce the use of region-proposing stage which is used to filter a majority of back-grounds for achieving real-time rates. Though foregrounds as hard examples are in urgent need of being mined from tons of backgrounds, a considerable number of state-of-the-art real-time detectors, like YOLO series, have yet to profit from existing hard example mining methods, as using these methods need detectors fit series of prerequisites. In this paper, we propose a general hard example mining method named Loss Rank Mining (LRM) to fill the gap. LRM is a general method for real-time detectors, as it utilizes the final feature map which exists in all real-time detectors to mine hard examples. By using LRM, some elements representing easy examples in final feature map are filtered and detectors are forced to concentrate on hard examples during training. Extensive experiments validate the effectiveness of our method. With our method, the improvements of YOLOv2 detector on auto-driving related dataset KITTI and more general dataset PASCAL VOC are over 5% and 2% mAP, respectively. In addition, LRM is the first hard example mining strategy which could fit YOLOv2 perfectly and make it better applied in series of real scenarios where both real-time rates and accurate detection are strongly demanded. © 2018 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Grid computing in China

引用

Journal of Grid Computing 2004年第2期2卷 193-206页

作者： Yang, Guangwen Jin, Hai Li, Minglu Xiao, Nong Li, Wei Wu, Zhaohui Wu, Yongwei Tang, Feilong Department of Computer Science and Technology Tsinghua University Beijing China School of Computer Huazhong University of Science and Technology Wuhan China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Key Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha China Institute of Computing Technology Chinese Academy of Sciences Beijing 100080 China College of Computer Science Zhejiang University Hangzhou 310027 China

Grid computing presents a new trend to distributed computation and Internet applications, which can construct a virtual single image of heterogeneous resources, provide uniform application interface and integrate widespread computational resources into super, ubiquitous and transparent aggregation. In the adoption of Grid computing, China, who is facing more resource heterogeneity and other specific demands, has put much effort on both research and practical utilization. In this paper, we introduce the major China Grid research projects and their perspective applications. First we give the overview of the four government-sponsored programs in Grid, namely the China National Grid, ChinaGrid, NSFC Grid, and ShanghaiGrid. Then we present six representative ongoing Grid systems in details, which are categorized into Grid middleware and Grid application. This paper provides the general picture of Grid computing in China, and shows the great efforts, devotion and confidence in China to use Grid technology to boost the society, economics and scientific research. © Springer 2004.

关键词： Grid computing

来源：评论

学校读者我要写书评

暂无评论

Integrated high-speed optical SerDes over 100GBd based on optical time division multiplexing

Integrated high-speed optical SerDes over 100GBd based on op...

引用

作者： Xu, Shi Luo, Zhang Lai, Mingche Pang, Zhengbin Li, Renfa College of Computer Hunan University Changsha410082 China College of Computer National University of Defense Technology Changsha410073 China Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha410073 China

An on-chip optical transceiver for transmission system over 100GBd is proposed based on optical time division multiplexing (OTDM) technology, and the performances, such as the insertion loss, the inter-symbol interference (ISI) crosstalk, and the potential symbol rate, are analyzed in detail. Co-designed with the double rail driver, on-chip Mach-Zehnder interferometer switch repeatedly generates extremely narrow sampling pulses of only 12ps full width at half maximum. Based on such narrow optical sampling pulse train, a four-stage cascaded optical switch divides the 25GHz clock cycle into four recurrent 9.5ps time slots and one blank time slot of 2ps. Thus, a 100GBd optical transmission channel is realized based on 4-bit 25Gbps bit-streams at the electrical interface. The ISI extinction ratio at the worst channel is 1.9dB with 10dB depth modulator, and the insertion loss caused by the OTDM mechanism is about 16dB. Further, taking advantages of dark modulation, an OTDM system with 5-bit 25Gbps bit-streams at the electrical interface is proposed to generate a 125GBd transmission utilizing the same optical sampling pulse. The ISI performance is much better and the extinction ratio at the worst channel is enhanced to 3.99dB. © 2018 ACM.

关键词： Timing circuits

来源：评论

学校读者我要写书评

暂无评论

Simultaneously Learning Syntactic Dependency and Semantics Reasonability for Relation Extraction

Simultaneously Learning Syntactic Dependency and Semantics R...

引用

International Conference on Image, Vision and Intelligent Systems, ICIVIS 2021

作者： Wang, Xin Yin, Nan Zhang, Xiang Bai, Xinyi Luo, Zhigang College of Computer National University of Defense Technology Changsha China Institute for Quantum and State Key Laboratory of High Performance Computing National University of Defense Technology Changsha China Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China

ISBN: (纸本)9789811669620

Relation extraction as an important Natural Language Processing (NLP) task is to identify relations between named entities in text. Recently, graph convolutional networks over dependency trees have been widely used to capture syntactic features and achieved attractive performance. However, most existing dependency-based approaches ignore the positive influence of the words outside the dependency trees, sometimes conveying rich and useful information on relation extraction. In this paper, we propose a novel model, Entity-aware Self-attention Contextualized GCN (ESC-GCN), which efficiently incorporates syntactic structure of input sentences and semantic context of sequences. To be specific, relative position self-attention obtains the overall semantic pairwise correlation related to word position, and contextualized graph convolutional networks capture rich intra-sentence dependencies between words by adequately pruning operations. In this way, our proposed model not only reduces the noisy impact from dependency trees but also obtains easily-ignored entity-related semantic representation. Extensive experiments demonstrate that our model achieves encouraging performance. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Simulation research of communication networks based on prim algorithm

Simulation research of communication networks based on prim ...

引用

International Forum on Information technology and Applications

作者： Yang, Chenghui Ren, ChuanJun College of Electrical Engineering Northwest University for Nationalities Lanzhou 730030 China National Laboratory for Parallel and Distributed Processing College of Computer Science National University of Defense Technology Changsha 410073 China

ISBN: (纸本)9780769541150

In order to improve the efficiency of the communication networks, we used the Kruskal algorithm and the Prim algorithm through algorithm comparison and analysis methods of data structure. A dynamic framework for the communication network installed system is built. Moreover, according to the actual framework of the communication network specific issues, the module chooses the Minimum Cost Spanning Tree Prim algorithm ultimately. The assumptions process and outcomes simulation have proper analysis and certification by C language. Experimental results show that the Minimum Cost Spanning Tree of Prim algorithm on the installed system for communication networks dynamic planning process have the advantage of speed and effectively reduce the waste of resources, which not only can ensure efficiency but also can effectively improve communication networks installed cost. © 2010 IEEE.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

Multi-Outputs Is All You Need For Deblur

arXiv

引用

arXiv 2022年

作者： Liu, Sidun Qiao, Peng Dou, Yong Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Hunan China

Image deblurring task is an ill-posed one, where exists infinite feasible solutions for blurry image. Modern deep learning approaches usually discard the learning of blur kernels and directly employ end-to-end supervised learning. Popular deblurring datasets define the label as one of the feasible solutions. However, we argue that it’s not reasonable to specify a label directly, especially when the label is sampled from a random distribution. Therefore, we propose to make the network learn the distribution of feasible solutions, and design based on this consideration a novel multi-head output architecture and corresponding loss function for distribution learning. Our approach enables the model to output multiple feasible solutions to approximate the target distribution. We further propose a novel parameter multiplexing method that reduces the number of parameters and computational effort while improving performance. We evaluated our approach on multiple image-deblur models, including the current state-of-the-art NAFNet. The improvement of best overall (pick the highest score among multiple heads for each validation image) PSNR outperforms the compared baselines up to 0.11∼0.18dB. The improvement of the best single head (pick the best-performed head among multiple heads on validation set) PSNR outperforms the compared baselines up to 0.04∼0.08dB. The codes are available at https://***/Liu-SD/multi-output-deblur. © 2022, CC BY.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Diagonalwise Refactorization: An Efficient Training Method for Depthwise Convolutions

Diagonalwise Refactorization: An Efficient Training Method f...

引用

International Joint Conference on Neural Networks

作者： Zheng Qin Zhaoning Zhang Dongsheng Li Yiming Zhang Yuxing Peng Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China

Depthwise convolutions provide significant performance benefits owing to the reduction in both parameters and mult-adds. However, training depthwise convolution layers with GPUs is slow in current deep learning frameworks because their implementations cannot fully utilize the GPU capacity. To address this problem, in this paper we present an efficient method (called diagonalwise refactorization) for accelerating the training of depthwise convolution layers. Our key idea is to rearrange the weight vectors of a depthwise convolution into a large diagonal weight matrix so as to convert the depthwise convolution into one single standard convolution, which is well supported by the cuDNN library that is highly-optimized for GPU computations. We have implemented our training method in five popular deep learning frameworks. Evaluation results show that our proposed method gains 15.4× training speedup on Darknet, 8.4× on Caffe, 5.4× on PyTorch, 3.5× on MXNet, and 1.4× on TensorFlow, compared to their original implementations of depthwise convolutions.

关键词： Standards Graphics processing units Kernel Convolutional codes Training Acceleration Libraries

来源：评论

学校读者我要写书评

暂无评论

Delayed Switch:Cloud Service Upgrade with Low Availability and Capacity Loss

Delayed Switch:Cloud Service Upgrade with Low Availability a...

引用

2014 IEEE 5th International Conference on Software Engineering and Service science

作者： Xue Ouyang Bo Ding Huaimin Wang Parallel and Distributed Laboratory School of Computer Science National University of Defense Technology

Cloud services must upgrade continuously in order to maintain ***,a large body of empirical evidence suggests that,upgrade procedures used in practice are failure-prone and often cause planned or unplanned *** this paper,we first define what is cloud service online upgrade,and then we analyze the shortcomings of current mainstream cloud service online upgrade *** particular,the mixed version problem along with rolling upgrade and the capacity loss problem brought by split mode upgrade have been *** that,we propose a solution called Delayed Switch,which can conduct cloud service upgrade with lower loss of availability and capacity in contrast with existing *** prove the performance of delayed switch in theory and develop a prototype system applying this *** by conducting experiments with a typical e-commerce service named Rubis,we validate the effectiveness and efficiency of our approach.

关键词： cloud service availability on-line upgrade mixed version problem split mode upgrade

来源：评论

学校读者我要写书评

暂无评论

The parallelized Cuckoo Filter for Cold Data Representation 23

The Parallelized Cuckoo Filter for Cold Data Representation

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Sun, Bowen Luo, Lailong Li, Shangsen Chen, Yingwen Guo, Deke College of Computer National University of Defense Technology China National University of Defense Technology Science and Technology on Information Systems Engineering Laboratory China College of Computer National University of Defense Technology National Lab for Parallel and Distributed Processing China

ISBN: (纸本)9781665494571

Cold data contributes a large portion of the big data today and is usually stored in secondary storage. Various sketch data structures are implemented to represent the stored elements and provide constant-time membership queries. Considering the large volume and infrequent access of cold data, we envision three rationales of cold data representation: 1) parallelized insertion, 2) accurate representation, and 3) constant-time query/deletion. The existing sketches, however, fail to achieve these rationales simultaneously. To this end, we propose a novel sketch named parallelized cuckoo Filter (PCF) in this paper. PCF divides the standard cuckoo filter into multiple homogenous segments. An additional hash function is introduced to assign an element to a segment. The element will be inserted into the selected segment. With such a design, element insertions in PCF can be naturally parallelized. Besides, as a variant of the cuckoo filter, PCF naturally realizes the accurate representation and constant-time query/deletion requirements. Comprehensive experiments further indicate that PCF achieves the three rationales simultaneously. © 2021 IEEE.

关键词： Hash functions Smart cities Big Data Data structures Standards

来源：评论

学校读者我要写书评

暂无评论

Predicting potential gene ontology from cellular response data 17

Predicting potential gene ontology from cellular response da...

引用

5th International Conference on Bioinformatics and Computational Biology, ICBCB 2017

作者： Hong, Hao Yin, Xiaoyao Li, Fei Guan, Naiyang Bo, Xiaochen Luo, Zhigang Department of Chemistry and Biology National University of Defense Technology Changsha China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha China Department of Biotechnology Beijing Institute of Radiation Medicine Beijing China

ISBN: (纸本)9781450348270

Ontologies have proven to be useful for capturing and organizing knowledge as a hierarchical set of terms and their relationships. However, curating gene ontology data by hand requires specialized knowledge of certain field, which is inefficient. Thus inferring gene ontology from the exponentially increased biological data is getting hot. Based on the Library of Integrated Network-Based Cellular Signatures (LINCS) data we came up with the hypothesis that genes participate in analogous biological processes might affect cells accordantly. By assessing cellular response after genes were knock out we built a similarity matrix with the Gene Set Enrichment Analysis (GSEA) and clustered the genes with affinity propagation algorithm. Next we mapped the cluster result to gene ontology biological process data for annotation and enrichment analysis, which confirmed our hypothesis and made it possible to predict biological processes for unannotated genes from cellular response data after genes are knock out for the first time. We further validated the rationality from the gene ontology molecular function data. © 2017 ACM.

关键词： Gene Ontology

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：