检索结果-内蒙古大学图书馆

Knowledge graph embeddings based on 2d convolution and self-attention mechanisms for link prediction

APPLIEd INTELLIGENCE 2025年第2期55卷 1-18页

作者： Zan, Shijie Ji, Weidong Zhou, Guohui Harbin Normal Univ Coll Comp Informat Engn Harbin 150025 Peoples R China

Link prediction refers to using existing facts in the knowledge graph to predict missing facts. This process can enhance the integrity of the knowledge graph and facilitate various downstream applications. However, existing link prediction models usually extract features only in a global or local scope, resulting in feature extraction being limited to a single scope. Additionally, to achieve optimal results, many models require increasing embedding dimensions and parameter numbers, which can lead to scalability issues when applied to large knowledge graphs. To address these issues, we propose a model that fuses the self-attention mechanism with 2d convolution for the link prediction task. The model utilizes a self-attention mechanism with numerous heads to capture feature interactions between entities and relations in the global scope. Furthermore, we innovatively introduce 2d convolution to capture feature interactions in the local scope. Results using FB15k-237 and WN18RR as standard link prediction benchmarks show that our fusion model has good comparable performance with current state-of-the-art models. In particular, compared to the ConvE model (which uses only 2d convolution), our proposed model achieves 13.7% and 14.7% improvement in MRR metrics, and compared to the SAttLE model (which uses only the self-attention mechanism) achieves 2.5% and 0.5% improvement in MRR metrics. Furthermore, due to the low-dimensional embedding of entities and relations, our proposed model has low complexity, good scalability, and thus can accomplish link prediction tasks on larger knowledge graphs in the real world.

关键词： Knowledge graph Low-dimensional embedding 2d convolution Multi-head self-attention mechanism Feature interaction R-drop structure

来源：评论

学校读者我要写书评

暂无评论

Crime amount prediction based on 2d convolution and long short-term memory neural network

引用

ETRI JOURNAL 2022年第2期44卷 208-219页

作者： dong, Qifen Ye, Ruihui Li, Guojun Zhejiang Police Coll Dept Comp & Informat Secur Hangzhou Peoples R China Zhejiang Coll Secur Technol Dept Intelligent Secur Wenzhou Peoples R China Zhejiang Police Coll Dept Basic Courses Hangzhou Peoples R China

Crime amount prediction is crucial for optimizing the police patrols' arrangement in each region of a city. First, we analyzed spatiotemporal correlations of the crime data and the relationships between crime and related auxiliary data, including points-of-interest (POI), public service complaints, and demographics. Then, we proposed a crime amount prediction model based on 2d convolution and long short-term memory neural network (2dCONV-LSTM). The proposed model captures the spatiotemporal correlations in the crime data, and the crime-related auxiliary data are used to enhance the regional spatial features. Extensive experiments on real-world datasets are conducted. Results demonstrated that capturing both temporal and spatial correlations in crime data and using auxiliary data to extract regional spatial features improve the prediction performance. In the best case scenario, the proposed model reduces the prediction error by at least 17.8% and 8.2% compared with support vector regression (SVR) and LSTM, respectively. Moreover, excessive auxiliary data reduce model performance because of the presence of redundant information.

关键词： 2d convolution crime amount prediction LSTM temporal-spatial correlations regional spatial features

来源：评论

学校读者我要写书评

暂无评论

Exploration of OpenCL 2d convolution Kernels on Intel FPGA, CPU, and GPU Platforms

Exploration of OpenCL 2D Convolution Kernels on Intel FPGA, ...

引用

IEEE International Conference on Big data (Big data)

作者： Jin, Zheming Finkel, Hal Argonne Natl Lab Leadership Comp Facil Lemont IL 60439 USA

ISBN: (纸本)9781728108582

There is a need to evaluate the resource usage and optimize the performance of the floating-point 2d convolution kernels on a recent FPGA which features large numbers of hardened floating-point digital signal processing blocks and an increasingly large on-chip memory. In this paper, we presented an OpenCL 2d convolution kernel with configurable parameters for specifying the precision, sizes of filter and block, vectorization width, and compute-unit duplication factor. Then, we instantiated a set of specific instances of the kernel with a fixed filter size to evaluate their resource usage and performance to narrow down the exploration space. Based on the evaluation results on an Intel (R) Arria 10 FPGA using high-level synthesis, we evaluated the kernels with different filter sizes within the pruned exploration space. Compared to the baseline implementation in which the vectorization width is two and the block size is 32x32, our optimizations improve the performance by a factor ranging from 1.9X to 3X for the single-precision kernels, and from 2.2X to 3.37X for the half-precision kernels. Furthermore, we evaluated the performance and power of the kernels on an Intel (R) Xeon (R) CPU and an Iris (TM) Pro integrated GPU. We found that the FPGA could achieve the highest performance for a 9x9 filter among the CPU, GPU, and FPGA, but the GPU can achieve the highest performance fur other filter sizes.

关键词： OpenCL 2d convolution Floating Point FPGA CPU GPU

来源：评论

学校读者我要写书评

暂无评论

A design methodology for performance-resource optimization of a generalized 2d convolution architecture with quadrant symmetric kernels

A design methodology for performance-resource optimization o...

引用

12th Asia-Pacific Computer Systems Architecture Conference (ACSAC 2007)

作者： Zhang, Ming Z. Asari, Vijayan K. Old Dominion Univ Dept Elect & Comp Engn Computat Intelligence & Machine Vis Lab Norfolk VA 23529 USA

ISBN: (纸本)9783540743088

We present a design technique to meet application driven constraints for performance-resource optimization of a generalized 2d convolution with quadrant symmetric kernels. A fully pipelined multiplierless digital architecture for computing modularized 2d convolution utilizing the quadrant symmetry of the kernels is proposed in this paper. Pixels in the four quadrants of the kernel region with respect to an image pixel are considered simultaneously with distributed queues for computing the partial results of the convolution sum in time-sliced fashion. The new architecture performs computations in log-domain by utilizing low complexity high performance log(2) and inverse-log(2) estimation modules. An effective data handling strategy is developed to minimize routing of data path in conjunction with the logarithmic modules to eliminate the necessity of the multipliers in the architecture. The proposed architecture is capable of performing convolution operations for 45.5 (1024x1024) frames or 47.73 million outputs per second in minimum resource configuration with 8x8 kernels in a Xilinx's Virtex XC2V2000-4ff896 field programmable gate array (FPCA) at maximum clock frequency of 190.92 MHz. Analysis shows that the performance and resource utilization between the fully parallel and fully resource constrained architectures are proportional to f and 1/f, respectively where f is the application driven reusability of the main computing components. In addition to resource reduction from optimization scheme, evaluation in Xilinx's core generator also showed that the proposed design results in 60% reduction in hardware when compared to the design using pipelined multipliers.

关键词： 2d convolution log-domain computation multiplier-less architecture quadrant symmetric kernels modularized optimization FPGA based architecture

来源：评论

学校读者我要写书评

暂无评论

High Performance Implementation of 2d convolution using Intel's Advanced Vector Extensions 19

High Performance Implementation of 2D Convolution using Inte...

引用

19th CSI International Symposium on Artificial Intelligence and Signal Processing (AISP)

作者： Amiri, Hossein Shahbahrami, Asadollah Univ Guilan Fac Engn Dept Comp Engn Rasht Iran

ISBN: (纸本)9781538625859

convolution is the most important and fundamental concept in multimedia processing. For example, for digital image processing 2d convolution is used for different filtering operations. It has many mathematical operations and is performed on all image pixels. Therefore, it is almost a compute-intensive kernel. In order to improve its performance in this paper, we apply two approaches to vectorize it, broadcasting of coefficients and repetition of coefficients using Intrinsic Programming Model (IPM) and AVX technology. Our experimental results on an Intel Skylake microarchitecture show that the performance of broadcasting of coefficients is much higher than repetition of coefficients for different filter sizes and different image sizes. In addition, in order to evaluate the performance of Compiler Automatic Vectorization (CAV), and OpenCV library for this kernel, we use GCC and LLVM compilers. Our experimental results show that the performance of both IPM implementations are faster than GCC's and LLVM auto-vectorizations.

关键词： Parallel Programming 2d convolution Vectorization SIMdization AVX2

来源：评论

学校读者我要写书评

暂无评论

dynamically Reconfigurable Parallel Architecture Implementation of 2d convolution for Image Processing over FPGA 2

Dynamically Reconfigurable Parallel Architecture Implementat...

引用

2nd International Conference on Electrical Engineering and Information and Communication Technology (ICEEICT)

作者： Jahiruzzaman, Md. Saha, Shumit Hawlader, Md. Abul Khayum Khulna Univ Engn & Technol Dept Elect & Commun Engn Khulna 9203 Bangladesh

ISBN: (纸本)9781467366762

2d convolution is the most convenient method to analyze digital image and video processing and has a variety of digital image processing applications such as edge detection, image enhancement, image segmentation, smoothing or blurring an image and can be applied to video processing on the basis of frame by frame for motion detection. Though the computational complexity of 2d convolution is comparatively high as it demands high level parallelism both for product and addition operations, it can be implemented on real time embedded system applications such as Application-Specific Integrated circuit (ASIC) or Field Programmable Gate Array (FPGA). As the convolved image directly depends on the kernel, proposed architecture is suitable for any kernel of 3x3 size. The objective of this study is to implement and synthesize a FPGA base image processing system based on 2d convolution on the basis of microprocessor architecture Simple-as-Possible. This system was designed and simulated in Verilog Hardware description Language (Verilog HdL) and synthesized on Virtex-5 FPGA. The result shows that, the proposed design has lesser number of blocks (look up tables & registers) than other architecture and low time delay. So that, the complexity of the proposed architecture is less and it can be used in image processing applications.

关键词： 2d convolution Verilog HdL FPGA Image Processing

来源：评论

学校读者我要写书评

暂无评论

Neighborhood dependent Approach for Low Power 2d convolution in Video Processing Applications

Neighborhood Dependent Approach for Low Power 2D Convolution...

引用

4th IEEE Conference on Industrial Electronics and Applications

作者： Ngo, Hau Asari, Vijayan USN Acad Dept Elect & Comp Engn Annapolis MD 21402 USA Old Dominion Univ Elect & Comp Engn Dept Norfolk VA USA

ISBN: (纸本)9781424427994

Window-based operations such as two dimensional (2-d) convolution operations are commonly used in image and video processing applications. In this paper, a new design technique that considers the neighboring pixels within the window to detect and eliminate redundant or unnecessary computations for power reduction is presented. A novel on-chip detection technique is developed for the proposed neighborhood dependent approach (NdA) to reduce computations. In addition, data partitioning methodology is employed in the on chip buffer design support real-time operations. This NdA method is applied to different window buffering schemes and experimental results are presented.

关键词： window-based operation FPGA neighborhood dependent approach 2d convolution 2d filter low power design

来源：评论

学校读者我要写书评

暂无评论

A fast stereo matching network based on temporal attention and 2d convolution

引用

PATTERN RECOGNITION 2023年第1期144卷

作者： Zhao, Youchen Zhong, Hua Jia, Boyuan Li, Haixiong Xidian Univ Sch Elect Engn 2 South Taibai Rd Xian 710071 Shaanxi Peoples R China

We propose a fast stereo matching network based on temporal attention and 2d convolution (TANet). due to the high similarity of the disparity between consecutive frames in an image sequence, we propose a temporal attention (TA) module that uses the disparity map of the previous frame to guide the disparity search range in the current frame, thus significantly improving the efficiency of disparity calculation in the cost volume module. Additionally, we propose a hierarchical cost construction and 2d convolution aggregation module that constructs a pyramid cost volume by fusing edge cues to establish detail constraints. This overcomes the problem of difficult convergence caused by information loss when replacing 3d convolution with 2d convolution. Experimental results show that the TA module effectively optimizes the cost volume and, together with 2d convolution, improves the computational speed. Compared with state-of-the-art algorithms, TANet achieves a speedup of nearly 4x, with a running time of 0.061s, and reduces the parameter count by nearly half while decreasing accuracy by 1.1%. Code is available at https://***/Y0uchenZ/TANet.

关键词： Stereo matching Temporal attention 2d convolution Edge cues Pyramid cost volume

来源：评论

学校读者我要写书评

暂无评论

An efficient implementation of 2d convolution in CNN

引用

IEICE ELECTRONICS EXPRESS 2017年第1期14卷 20161134-20161134页

作者： Chang, Jing Sha, Jin Nanjing Univ Sch Elect Sci & Engn Nanjing 210046 Jiangsu Peoples R China

convolutional neural network (CNN), a well-known machine learning algorithm, has been widely used in the field of computer vision for its amazing performance in image classification. With the rapid growth of applications based on CNN, various acceleration schemes have been proposed on FPGA, GPU and ASIC. In the implementation of these specific hardware accelerations, the most challenging part is the implementation of 2d convolution. To obtain a more efficient design of 2d convolution in CNN, this paper proposes a novel technique, singular value decomposition approximation (SVdA) to reduce the usage of resources. Experimental results show that the proposed SVdA hardware implementation can achieve a reduction in resources in the range of 14.46% to 37.8%, while the loss of classification accuracy is less than 1%.

关键词： CNN 2d convolution hardware implementation

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral image classification via principal component analysis, 2d spatial convolution, and support vector machines

引用

JOURNAL OF APPLIEd REMOTE SENSING 2021年第3期15卷

作者： Chen, Guang Y. Xie, Wenfang Krzyzak, Adam Qian, Shen-En Concordia Univ Dept Comp Sci & Software Engn Montreal PQ Canada Concordia Univ Dept Mech Ind & Aerosp Engn Montreal PQ Canada Canadian Space Agcy Space Sci & Technol St Hubert PQ Canada

Hyperspectral image (HSI) classification has many applications in different diverse research fields. We propose a method for HSI classification using principal component analysis (PCA), 2d spatial convolution, and support vector machine (SVM). Our method takes advantage of correlation in both spatial and spectral domains in an HSI data cube at the same time. We use PCA to reduce the dimensionality of an HSI data cube. We then perform spatial convolution to the dimension-reduced data cube once and then to the convolved data cube for the second time. As a result, we have generated two convolved PCA output data cubes in a multiresolution way. We feed the two convolved data cubes to SVM to classify each pixel to one of the known classes. Experiments on three widely used hyperspectral data cubes (i.e., Indian Pines, Pavia University, and Salinas) demonstrate that our method can improve the classification accuracy significantly when compared to a few existing methods. Our method is relatively fast in terms of central processing unit computational time as well. (C) 2021 Society of Photo-Optical Instrumentation Engineers (SPIE)

关键词： hyperspectral image classification 2d convolution support vector machines

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：