检索结果-内蒙古大学图书馆

Efficient Time Series Clustering and Its Application to Social Network Mining

JOURNAL OF INTELLIGENT SYSTEMS 2014年第2期23卷 213-229页

作者： Zhou, Cangqi Zhao, Qianchuan Tsinghua Univ Ctr Intelligent & Networked Syst Dept Automat Tsinghua Natl Lab Informat Sci & Technol Beijing 100084 Peoples R China

Mining time series data is of great significance in various areas. To efficiently find representative patterns in these data, this article focuses on the definition of a valid dissimilarity measure and the acceleration of partitioning clustering, a common group of techniques used to discover typical shapes of time series. Dissimilarity measure is a crucial component in clustering. It is required, by some particular applications, to be invariant to specific transformations. The rationale for using the angle between two time series to define a dissimilarity is analyzed. Moreover, our proposed measure satisfies the triangle inequality with specific restrictions. This property can be employed to accelerate clustering. An integrated algorithm is proposed. The experiments show that angle-based dissimilarity captures the essence of time series patterns that are invariant to amplitude scaling. In addition, the accelerated algorithm outperforms the standard one as redundancies are pruned. Our approach has been applied to discover typical patterns of information diffusion in an online social network. Analyses revealed the formation mechanisms of different patterns.

关键词： Time series clustering dissimilarity measure algorithm acceleration social network mining

来源：评论

学校读者我要写书评

暂无评论

Fast bit gather, bit scatter and bit permutation instructions for commodity microprocessors

引用

JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY 2008年第1-2期53卷 145-169页

作者： Hilewitz, Yedidya Lee, Ruby B. Princeton Univ PALMS Dept Elect Engn Princeton NJ 08544 USA

Advanced bit manipulation operations are not efficiently supported by commodity word-oriented microprocessors. Programming tricks are typically devised to shorten the long sequence of instructions needed to emulate these complicated bit operations. As these bit manipulation operations are relevant to applications that are becoming increasingly important, we propose direct support for them in microprocessors. In particular, we propose fast bit gather (or parallel extract), bit scatter (or parallel deposit) and bit permutation instructions (including group, butterfly and inverse butterfly). We show that all these instructions can be implemented efficiently using both the fast butterfly and inverse butterfly network datapaths. Specifically, we show that parallel deposit can be mapped onto a butterfly circuit and parallel extract can be mapped onto an inverse butterfly circuit. We define static, dynamic and loop invariant versions of the instructions, with static versions utilizing a much simpler functional unit. We show how a hardware decoder can be implemented for the dynamic and loop-invariant versions to generate, dynamically, the control signals for the butterfly and inverse butterfly datapaths. The simplest functional unit we propose is smaller and faster than an ALU. We also show that these instructions yield significant speedups over a basic RISC architecture for a variety of different application kernels taken from applications domains including bioinformatics, steganography, coding, compression and random number generation.

关键词： bit manipulations permutations bit scatter bit gather parallel extract parallel deposit pack unpack microprocessors instruction set architecture ISA algorithm acceleration bioinformatics pattern matching compression steganography cryptology

来源：评论

学校读者我要写书评

暂无评论

Parallel algorithm for Numerical Modeling of Anisotropic Heat and Mass Transfer in Fractal Media 17

Parallel Algorithm for Numerical Modeling of Anisotropic Hea...

引用

IEEE 17th International Conference on the Experience of Designing and Application of CAD Systems (CADSM)

作者： Sokolovskyy, Yaroslav Yarkun, Volodymyr Levkovych, Mariana Lviv Polytechn Natl Univ Dept Comp Aided Design Lvov Ukraine Ukrainian Natl Forestry Univ Dept Informat Technol Lvov Ukraine

ISBN: (纸本)9798350310856

This article considers the algorithmic aspects of parallelization of the numerical implementation of the mathematical model of anisotropic heat and mass transfer, obtained on the basis of the use of fractional order derivatives. Software for implementing a parallel algorithm for modeling heat and mass transfer in media with a fractal structure is presented. The acceleration indicators of the parallel algorithm with different discretization parameters were studied on the basis of numerical experiments. A parallel implementation was also compared with a serial solution. The proposed parallel algorithms and their software implementation can scale well on multi-core computers.

关键词： heat and mass transfer media with a fractal structure derivatives of fractional order parallel algorithm algorithm acceleration

来源：评论

学校读者我要写书评

暂无评论

Green Deep Learning: Comparative Study of Road Object Detectors Between Jetson Boards and PC 6th

Green Deep Learning: Comparative Study of Road Object Detect...

引用

6th Iberian Robotics Conference (Robot)

作者： Sanchez-Garcia, Fabio Arango, Felipe Gomez-Huelamo, Carlos Ocana, Manuel Revenga, Pedro Bergasa, Luis M. Univ Alcala Dept Elect RobeSafe Res Grp Madrid Spain

ISBN: (纸本)9783031586750;9783031586767

Recent advancements in deep learning have provided powerful tools for intelligent vehicle tasks, particularly in the field of perception. However, achieving real-time performance with low power consumption remains a challenge included in the hot research topic known as green deep learning. In this paper, we present a comparative analysis of various YOLOv5 weights trained on the KITTI and SHIFT datasets using two platforms with different power consumption profiles: the NVIDIA Jetson AGX Xavier and a desktop computer equipped with a NVIDIA GTX 1080 Ti. Our analysis focuses on the average inference time and precision metrics for road objects detection, a key task for intelligent vehicles. Additionally, we apply TensorRT to optimize and accelerate the architecture on both platforms, resulting in significant speed improvements, particularly on the low-power Jetson AGX Xavier (30W). Our ultimate goal is to implement our whole autonomous driving architecture on several Jetson AGX Xaviers connected to a PC where the hyper-realistic CARLA simulator, is replicating the real-world autonomous vehicle environment. We obtain compelling validation results on KITTI and CARLA, achieving real-time performance on a lightweight Jetson AGX Xavier with a powerful object detector such as YOLOv5m.

关键词： green deep learning algorithm acceleration embedded systems TensorRT

来源：评论

学校读者我要写书评

暂无评论

Increasing Network Size and Training Throughput of FPGA Restricted Boltzmann Machines using Dropout 24

Increasing Network Size and Training Throughput of FPGA Rest...

引用

24th IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM)

作者： Su, Jiang Thomas, David B. Cheung, Peter Y. K. Imperial Coll London Dept Elect & Elect Engeering London England

ISBN: (纸本)9781509023561

Restricted Boltzmann Machines (RBMs) are widely used in modern machine learning tasks. Existing implementations are limited in network size and training throughput by available DSP resources. In this work we propose a new algorithm and architecture for FPGAs called dropout-RBM (dRBM) system. Compared to the state-of-art design methods on the same FPGA, dRBM with a dropout rate 0.5 doubles the maximum affordable network size using only half of DSP and BRAM resources. This is achieved by an application of a technique called dropout, which is a relatively new method used to avoid overfitting of data. Here we instead apply dropout as a technique for reducing the required DSPs and BRAM resources, while also having the side-effect of increasing robustness of training. Also to improve the processing throughput, we propose a multi-mode matrix multiplication module that maximizes the DSP efficiency. For the MNIST classification benchmark, a Stratix IV EP4SGX530 FPGA running dRBM is 34x faster than a single-precision Matlab implementation running on Intel i7 2.9GHz CPU.

关键词： algorithm acceleration Dropout Restricted Boltzmann Machine

来源：评论

学校读者我要写书评

暂无评论

Joint qualitative and quantitative evaluation of fast image dehazing based on dark channel prior 3

Joint qualitative and quantitative evaluation of fast image ...

引用

Conference on Unconventional Optical Imaging III Part of SPIE Photonics Europe Conference

作者： Aksas, Lyes Lapray, Pierre-Jean Foulonneau, Alban Bigue, Laurent Georges, Marc P. Popescu, Gabriel Verrier, Nicolas Univ Haute Alsace EA7499 IRIMAS Mulhouse France

ISBN: (纸本)9781510651494;9781510651487

Haze is an undesirable effect in images caused when atmospheric particles, such as water droplets, ice crystals, dust, or smoke, are lit directly or indirectly by the sun. This effect can be counteracted by image processing, to bring back the details of a hazy image. Unfortunately, the execution time is often long, which prevents deployment in some video or real-time applications. In this paper, we propose to tune several parameters of the Dark Channel Prior method (DCP) algorithm combined with the fast guided filter. We evaluate the optimization in terms of execution time, and quantify the output image quality using different image quality metrics.

关键词： Dehazing Dark Channel Prior Optimization algorithm acceleration

来源：评论

学校读者我要写书评

暂无评论

Density-Based Fuzzy C-Means Multi-Center Re-Clustering Radar Signal Sorting algorithm 8

Density-Based Fuzzy C-Means Multi-Center Re-Clustering Radar...

引用

8th International Conference on Instrumentation and Measurement, Computer, Communication and Control (IMCCC)

作者： Cao, Sheng Wang, Shucheng Harbin Inst Technol Dept Microelect Sci & Technol Harbin 150001 Peoples R China Harbin Inst Technol Dept Informat Engn Harbin 150001 Peoples R China

ISBN: (纸本)9781538682463

y The strategic position of electronic warfare in modern warfare is constantly improving, and radar detection is the eye of modern information warfare and plays an important role in electronic warfare. This paper designs a new pulse radar sorting algorithm: a Density-Based Fuzzy C-Means Multi-Center Re-Clustering (DFCMRC) radar signal sorting algorithm. This algorithm combines the advantages of the DBSCAN density clustering algorithm and the fuzzy C-means (FCM) clustering algorithm. This paper also optimizes the structure of the DFCMRC algorithm, which changes the algorithm that randomly generated the initial center point to the CFSFDP algorithm. After comparison tests, the DFCMRC algorithm sorting result is better than the k-means algorithm, the DBSCAN algorithm and the FCM algorithm. Also, the membership degree description of DFCMRC is more reasonable than the FCM's. Accelerated optimized DFCMRC algorithm can reduce more than half iterations, which greatly shortens the algorithm calculation time.

关键词： radar signal sorting clustering algorithm algorithm acceleration

来源：评论

学校读者我要写书评

暂无评论

Accelerated Implementation of the Power Method for Analysis of Electromagnetic Fields 29

Accelerated Implementation of the Power Method for Analysis ...

引用

29th International Conference on Cybernetics & Informatics (K&I)

作者： Ruzicky, Eugen Stefanovic, Juraj Schindler, Frank Palenik, Tomas Pan European Univ Fac Informat Bratislava Slovakia Slovak Univ Technol Bratislava Fac Elect Engn & Informat Technol Bratislava Slovakia

ISBN: (纸本)9781538644218

This paper presents two kinds of optimizations to the well known power iteration method for finding dominant eigenvalue of matrices, with regard to special matrices often used in EM field measurement and processing. Based on the statistical analysis of actual measurement data, the first optimization suggests a method for improving the convergence of the power iteration algorithm. The second optimization is implementational: the acceleration of the power method by means of utilising modern vector processing features of the x86-64 architecture is evaluated. Various implementations using different parameters are evaluated by using the Intel MKL library, and a comparative analysis is presented for general and also specific matrices that are important in the EM field measurement and processing.

关键词： power method algorithm acceleration correlation matrix principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Performance of the Bitonic Mergesort Network on a Dataflow Computer

Performance of the Bitonic Mergesort Network on a Dataflow C...

引用

21st Telecommunications Forum (TELFOR)

作者： Rankovic, Vukasin Kos, Anton Tomazic, Saso Milutinovic, Veljko Univ Belgrade Sch Elect Engn Belgrade 11001 Serbia Univ Ljubljana Fac Elect Engn Ljubljana 61000 Slovenia

ISBN: (纸本)9781479914203

High speed computing and growing amounts of data are driving the quest for ever faster sorting algorithms. Sorting networks executing parallel sorting and dataflow computational paradigm are offered as a possible solution. In presented experiments Bitonic mergesort algorithm is implemented on an entry model of the Maxeler dataflow supercomputing system. Our results show, that sorting of a small size arrays on Maxeler, comparing to the fastest sorting algorithm on a CPU, achieves the speedup factor of 16. Using more advanced Maxeler systems, we expect to be able to sort larger arrays and achieve greater speedups.

关键词： dataflow computing bitonic mergesort parallel sorting algorithm acceleration sorting network

来源：评论

学校读者我要写书评

暂无评论

Density-Based Fuzzy C-Means Multi-Center Re-Clustering Radar Signal Sorting algorithm 17

Density-Based Fuzzy C-Means Multi-Center Re-Clustering Radar...

引用

17th IEEE International Conference on Machine Learning and Applications (IEEE ICMLA)

作者： Cao, Sheng Wang, Shucheng Zhang, Yan Harbin Inst Technol Dept Microelect Sci & Technol Harbin Heilongjiang Peoples R China Harbin Inst Technol Dept Informat Engn Harbin Heilongjiang Peoples R China Heilongjiang Univ Dept Informat Engn Harbin Heilongjiang Peoples R China

ISBN: (纸本)9781538668054

As the improving strategic position of electronic warfare in modern warfare, radar sorting detection becomes the eye of modern information warfare and plays an important role in it. This paper designs a new pulse radar sorting algorithm: a Density-Based Fuzzy C-Means Multi-Center Re-Clustering (DFCMRC) radar signal sorting algorithm. This algorithm mainly combines the advantages of Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm and Fuzzy C-means (FCM) clustering algorithm. This paper also optimizes the structure of the DFCMRC algorithm, which changes the algorithm that randomly generated the initial center point to the Clustering by Fast Search and Find of Density Peaks (CFSFDP) algorithm. After comparison tests, the DFCMRC algorithm sorting result is better than the K-means algorithm, the DBSCAN algorithm and the FCM algorithm. Also, the membership grade description of DFCMRC makes more sense than the FCM's. Accelerated optimized DFCMRC algorithm can reduce more than half iterations, which greatly shortens the algorithm calculation time.

关键词： radar signal sorting clustering algorithm algorithm acceleration

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：