检索结果-内蒙古大学图书馆

31st IEEE International Symposium on High Performance computer Architecture, HPCA 2025

作者： Zhai, Jiaqi Shi, Xuanhua Huang, Kaiyi Ye, Chencheng Hu, Weifang He, Bingsheng Jin, Hai Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Wuhan430074 China National University of Singapore School of Computing 119077 Singapore

ISBN: (纸本)9798331506476

In the realm of recommendation systems, achieving real-time performance in embedding similarity tasks is often hindered by the limitations of traditional Top-K sparse matrix-vector multiplication (SpMV) methods, which suffer from high latency due to inefficient memory access patterns. This paper identifies these critical gaps and introduces AccelES, a novel approach that significantly enhances the efficiency of Top-K SpMV. Our method employs a two-stage calculation scheme: the first stage utilizes a compact, low-bit dataset to quickly identify the most relevant entries, while the second stage performs full-precision calculations solely on this pruned subset, thereby minimizing computational overhead. Furthermore, AccelES incorporates innovative matrix representations, Ultra-CSR and Random-CSR, which optimize memory bandwidth utilization. Experimental results demonstrate that AccelES accelerates performance, surpassing state-of-the-art FPGA, GPU, and CPU solutions by factors of 3.4×, 2.5×, and 153.3×, respectively, under controlled conditions. These advancements not only enhance processing speed but also significantly improve real-time performance in recommendation systems, establishing AccelES as a pivotal contribution to the field of Top-K sparse matrix-vector multiplication. © 2025 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

LSGraph: A Locality-centric High-performance Streaming Graph Engine 24

LSGraph: A Locality-centric High-performance Streaming Graph...

引用

19th European Conference on computer Systems, EuroSys 2024

作者： Qi, Hao Wu, Yiyang He, Ligang Zhang, Yu Luo, Kang Cai, Minzhi Jin, Hai Zhang, Zhan Zhao, Jin National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computer Science University of Warwick United Kingdom Zhejiang Lab China

ISBN: (纸本)9798400704376

Streaming graph has been broadly employed across various application domains. It involves updating edges to the graph and then performing analytics on the updated graph. However, existing solutions either suffer from poor data locality and high computation complexity for streaming graph analytics, or need high overhead to search and move graph data to ensure ordered neighbors during streaming graph update. This paper presents a novel locality-centric streaming graph engine, called LSGraph, to enable efficient both graph analytics and graph update. The main novelty of this engine is a differentiated hierarchical indexed streaming graph representation approach to achieve efficient data search and movement for graph update and also maintain data locality and ordered neighbors for efficient graph analytics simultaneously. Besides, a locality-aware streaming graph data update mechanism is also proposed to efficiently regulate the distance of data movement, minimizing the overhead of memory access during graph update. We have implemented LSGraph and conducted a systematic evaluation on both real-world and synthetic datasets. Compared with three cutting-edge streaming graph engines, i.e., Terrace, Aspen, and PaC-tree, LSGraph achieves 2.98×-81.08×, 1.46×-12.56×, and 1.26×-10.31× speedups during graph update, while obtaining 1.02×-4.28×, 1.58×-3.55×, and 1.20×-2.72× speedups during graph analytics, respectively. © 2024 Owner/Author.

关键词： Engines

来源：评论

学校读者我要写书评

暂无评论

LGDB-Net: Dual-Branch Path for Building Extraction from Remote Sensing Image 30

LGDB-Net: Dual-Branch Path for Building Extraction from Remo...

引用

30th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2024

作者： Zhang, Ronghuan Zhao, Jing Li, Ming Zou, Qingzhi Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan China Shandong University of Traditional Chinese Medicine School of Intelligence and Information Engineering Jinan China

ISBN: (纸本)9798331515966

Extracting buildings from remote sensing images using deep learning techniques is a widely applied and crucial task. Convolutional Neural Networks (CNNs) adopt hierarchical feature representation, showcasing powerful capabilities in extracting local information but facing challenges in capturing global features. Transformers can address this limitation, but they perform poorly in extracting local features and significantly increase memory requirements and computational complexity. To overcome these challenges, we propose a method for building extraction from remote sensing images called LGDB-Net (Local-Global Dual-Branch Network), employing a dual-branch approach. Firstly, inspired by Swin Transformer, we designed GB-Former(Global Branch-Former) as the backbone network to model global information. We use a linear multi-head self-attention mechanism to reduce time and memory complexity while maintaining a large global receptive field. Additionally, we replace the traditional multi-layer perceptron with a convolution-enhanced multi-layer perceptron to improve channel feature representation, reduce model parameters, and enhance segmentation performance. Secondly, we use multiple Depth-wise Conv3×3 + LN (Layer Normalization) + GeLU (Gaussian Error Linear Unit) modules as the auxiliary branch for local detailed feature extraction. Finally, we adopt a multi-scale feature fusion strategy to integrate feature information from both branches. We conduct a series of experiments on three datasets: WHU, Massachusetts, and Inria. The experimental results demonstrate that the proposed method not only effectively improves segmentation accuracy with lower building omission and commission rates but also significantly reduces model parameters and computational complexity. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Spatiotemporal deformation analysis of glaciers and surrounding landscapes in the Shishapangma region using InSAR techniques

引用

Journal of Mountain science 2025年第1期22卷 210-229页

作者： LIANG Wenli LIU Rui JI Qin KOU Pinglang Chongqing Key Laboratory of GIS Application School of Geography and TourismChongqing Normal UniversityChongqing 401331China National Earth System Science Data Center Beijing 100000China Chongqing Engineering Research Center of Spatial Big Data Intelligent Technology Chongqing University of Posts and TelecommunicationsChongqing 400065China Key Laboratory of Tourism Multisource Data Perception and Decision Ministry of Culture and Tourism(TMDPDMCT)Chongqing University of Posts and TelecommunicationsChongqing 400065China

Glacier dynamics in the Himalayan midlatitudes,particularly in regions like the Shishapangma,are not yet fully understood,especially the localized topographic and climatic impacts on glacier *** study analyzes the spatiotemporal characteristics of glacier surface deformation in the Shishapangma region using the Small Baseline Subset(SBAS)Interferometric Synthetic Aperture Radar(In SAR)*** analysis reveals an average deformation rate of-4.02±17.65 mm/yr across the entire study area,with glacier regions exhibiting significantly higher rates of uplift(16.87±13.20 mm/yr)and subsidence(20.11±14.55 mm/yr)compared to non-glacier *** identifies significant surface lowering on the mountain flanks and localized uplift in certain catchments,emphasizing the higher deformation rates in glacial areas compared to non-glacial *** found a strong positive correlation between temperature and cumulative deformation(correlation coefficient of 0.63),particularly in glacier areas(0.82).The research highlights the role of temperature as the primary driver of glacier wastage,particularly at lower elevations,with strong correlations found between temperature and cumulative *** also indicates the complex interactions between topographic features,notably,slope gradient,which shows a positive correlation with subsidence rates,especially for slopes below 35°.South-,southwest-,and west-facing slopes exhibit significant uplift,while north-,northeast-,and east-facing slopes predominantly ***,we identified transition zones between debris-covered glaciers and clean ice as areas of most intense deformation,with average rates exceeding 30 mm/yr,highlighting these as potential high-risk zones for *** study comprehensively analyzes the deformation characteristics in both glacier and non-glacier areas in the Shishapangma region,revealing the complex interplay of topographic,climatic,and hydrological factors influencing glacier dynamic

关键词： Glacier melting SBAS-InSAR Topographic factors Temperature change Middle Himalayas

来源：评论

学校读者我要写书评

暂无评论

RankFIRST: Visual Analysis for Factor Investment By Ranking Stock Timeseries

引用

IEEE Transactions on Visualization and computer Graphics 2022年 PP卷 1-10页

作者： Guo, Huijie Liu, Meijun Yang, Bowen Sun, Ye Qu, Huamin Shi, Lei SKLSDE and Beijing Advanced Innovation Center for Big Data and Brain Computing School of Computer Science and Engineering Beihang University China Hong Kong University of Science and Technology Hong Kong

In the era of quantitative investment, factor-based investing models are widely adopted in the construction of stock portfolios. These models explain the performance of individual stocks by a set of financial factors, e.g., market beta and company size. In industry, open investment platforms allow the online building of factor-based models, yet set a high bar on the engineering expertise of end-users. State-of-the-art visualization systems integrate the whole factor investing pipeline, but do not directly address domain users' core requests on ranking factors and stocks for portfolio construction. The current model lacks explainability, which downgrades its credibility with stock investors. To fill the gap in modeling, ranking, and visualizing stock time series for factor investment, we designed and implemented a visual analytics system, namely RankFIRST. The system offers built-in support for an established factor collection and a cross-sectional regression model viable for human interpretation. A hierarchical slope graph design is introduced according to the desired characteristics of good factors for stock investment. A novel firework chart is also invented extending the well-known candlestick chart for stock time series. We evaluated the system on the full-scale Chinese stock market data in the recent 30 years. Case studies and controlled user evaluation demonstrate the superiority of our system on factor investing, in comparison to both passive investing on stock indices and existing stock market visual analytics tools. IEEE

关键词： Investments

来源：评论

学校读者我要写书评

暂无评论

Attention-Based Residual Dense Shrinkage Network for ECG Denoising

引用

computer Modeling in engineering & sciences 2024年第3期138卷 2809-2824页

作者： Dengyong Zhang Minzhi Yuan Feng Li Lebing Zhang Yanqiang Sun Yiming Ling Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha University of Science and TechnologyChangsha410114China School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha410114China School of Computer and Artificial Intelligence Huaihua UniversityHuaihua418000China School of Computer National University of Defense TechnologyChangsha410073China New Energy Centralized Control Center CHN Energy Hunan Power New Energy Co.Ltd.Changsha410000China

Electrocardiogram(ECG)signal is one of the noninvasive physiological measurement techniques commonly usedin cardiac ***,in real scenarios,the ECGsignal is susceptible to various noise erosion,which affectsthe subsequent pathological ***,the effective removal of the noise from ECG signals has becomea top priority in cardiac diagnostic *** at the problem of incomplete signal shape retention andlow signal-to-noise ratio(SNR)after denoising,a novel ECG denoising network,named attention-based residualdense shrinkage network(ARDSN),is proposed in this ***,the shallow ECG characteristics are extractedby a shallow feature extraction network(SFEN).Then,the residual dense shrinkage attention block(RDSAB)isused for adaptive noise ***,feature fusion representation(FFR)is performed on the hierarchicalfeatures extracted by a series of RDSABs to reconstruct the de-noised ECG *** on the MIT-BIHarrhythmia database and MIT-BIH noise stress test database indicate that the proposed scheme can effectively resistthe interference of different sources of noise on the ECG signal.

关键词： Electrocardiogram signal denoising signal-to-noise ratio attention-based residual dense shrinkage network MIT-BIH

来源：评论

学校读者我要写书评

暂无评论

A Parallel Color Image Encryption Algorithm Based on a 2-D Logistic-Rulkov Neuron Map

引用

IEEE Internet of Things Journal 2025年第11期12卷 18115-18124页

作者： Gao, Suo Zhang, Zheyi Iu, Herbert Ho-Ching Ding, Siqi Mou, Jun Erkan, Uğur Toktas, Abdurrahim Li, Qi Wang, Chunpeng Cao, Yinghong Dalian Polytechnic University School of Information Science and Engineering Dalian116034 China The University of Western Australia School of Electrical Electronic and Computer Engineering PerthWA6009 Australia Dalian University of Technology School of Mathematical Sciences Dalian116024 China Karamanoglu Mehmetbey University Faculty of Engineering Department of Computer Engineering Karaman70200 Turkey Ankara University Engineering Faculty Department of Artificial Intelligence and Data Engineering Ankara06830 Turkey Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan250000 China

Images are widely used in social networks, necessitating efficient and secure transmission, especially in bandwidth-constrained environments. This article aims to develop a color image encryption algorithm that enhances security while optimizing computational efficiency. A novel parallel color image encryption algorithm based on the 2-D logistic-Rulkov neuron map (2D-LRNM) is proposed. In this approach, the three channels of the color image are first separated. Cross-channel information interaction is introduced to form three new channels, which are then processed in parallel. During the encryption process of each channel, a block-wise parallel encryption mechanism is applied, ensuring simultaneous encryption of each block. This block-wise strategy effectively leverages parallel computing resources and balances the task load. To meet the demand for a large number of keystreams during encryption, the 2D-LRNM is introduced. It combines the simplicity and chaotic properties of the Logistic map with the multitimescale dynamics and neurodynamic behaviors of the Rulkov map. By overcoming the dimensional limitations inherent in the single Logistic map, this approach extends the system to a 2-D framework, significantly increasing the complexity of chaotic behavior and improving its unpredictability. Experimental results demonstrate that the proposed encryption algorithm achieves high security and reduces computation time by approximately 83.3%. © 2014 IEEE.

关键词： Encryption algorithms

来源：评论

学校读者我要写书评

暂无评论

Deep attention enhanced networks for medical image segmentation

Deep attention enhanced networks for medical image segmentat...

引用

2024 International Conference on Remote Sensing, Mapping, and Image Processing, RSMIP 2024

作者： Shen, Longfeng Wang, Qiong Wang, Wei Wang, Ran Yu, Yexun Zhang, Changchang School of Computer Science and Technology Huaibei Normal University Huaibei China Institute of Artificial lntelligence Hefei Comprehensive National Science Center Hefei China Anhui Big-Data Research Center on University Management Anhui Huaibei China

ISBN: (纸本)9781510680012

Medical image segmentation is a crucial task within the realm of medical image processing. Nevertheless, the intrinsic characteristics of medical images and the limited availability of data constrain the model's generalization capacity. Addressing this challenge requires an infusion of more data and the implementation of effective segmentation techniques to enhance model performance. In response to this need, we propose a deep attention enhanced network for medical image segmentation. This innovative approach boosts the segmentation model's efficacy through techniques such as data augmentation, a deep attention enhanced decoder, and a dual convolutional segmentation head. Validation across multiple datasets substantiates the method's effectiveness and its robust generalization capabilities. © 2024 SPIE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A Low-Rank Matching Attention Based Cross-Modal Feature Fusion Method for Conversational Emotion Recognition

引用

IEEE Transactions on Affective Computing 2024年第2期16卷 1177-1189页

作者： Shou, Yuntao Liu, Huan Cao, Xiangyong Meng, Deyu Dong, Bo Xi'An Jiaotong School of Computer Science and Technology Ministry of Education Key Laboratory for Intelligent Networks and Network Security Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Shaanxi Xi'an710049 China Xi'An Jiaotong School of Computer Science and Technology Shaanxi Xi'an710049 China BigKEJoint Innovation Center SHMEEA Shanghai China Xi'An Jiaotong University School of Mathematics and Statistics Xi'an China

Conversational emotion recognition (CER) is an important research topic in human-computer interactions. Although recent advancements in transformer-based cross-modal fusion methods have shown promise in CER tasks, they tend to overlook the crucial intra-modal and inter-modal emotional interaction or suffer from high computational complexity. To address this, we introduce a novel and lightweight cross-modal feature fusion method called Low-Rank Matching Attention Method (LMAM). LMAM effectively captures contextual emotional semantic information in conversations while mitigating the quadratic complexity issue caused by the self-attention mechanism. Specifically, by setting a matching weight and calculating inter-modal features attention scores row by row, LMAM requires only one-third of the parameters of self-attention methods. We also employ the low-rank decomposition method on the weights to further reduce the number of parameters in LMAM. As a result, LMAM offers a lightweight model while avoiding overfitting problems caused by a large number of parameters. Moreover, LMAM is able to fully exploit the intra-modal emotional contextual information within each modality and integrates complementary emotional semantic information across modalities by computing and fusing similarities of intra-modal and inter-modal features simultaneously. Experimental results verify the superiority of LMAM compared with other popular cross-modal fusion methods on the premise of being more lightweight. Also, LMAM can be embedded into any existing state-of-the-art CER methods in a plug-and-play manner, and can be applied to other multi-modal recognition tasks, e.g., session recommendation and humour detection, demonstrating its remarkable generalization ability. © 2024 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Joint Pre-Encoding Representation and Sturcture Embedding for Efficient and Low-Resource Knowledge Graph Completion

Joint Pre-Encoding Representation and Sturcture Embedding fo...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Qiu, Chenyu Qian, Pengjiang Wang, Chuang Yao, Jian Liu, Li Fang, Wei Ng, Eddie-Yin-Kwee School of Artificial Intelligence and Computer Science Jiangnan University China Engineering Research Center of Intelligent Technology for Healthcare Ministry of Education Jiangnan University China Big Data Center Affiliated Hospital of Jiangnan University China Nanyang Technological University Singapore

ISBN: (纸本)9798891761643

Knowledge graph completion (KGC) aims to infer missing or incomplete parts in knowledge graph. The existing models are generally divided into structure-based and description-based models, among description-based models often require longer training and inference times as well as increased memory usage. In this paper, we propose Pre-Encoded Masked Language Model (PEMLM) to efficiently solve KGC problem. By encoding textual descriptions into semantic representations before training, the necessary resources are significantly reduced. Furthermore, we introduce a straightforward but effective fusion framework to integrate structural embedding with pre-encoded semantic description, which enhances the model's prediction performance on 1-N relations. The experimental results demonstrate that our proposed strategy attains state-of-the-art performance on the WN18RR (MRR+5.4% and Hits@1+6.4%) and UMLS datasets. Compared to existing models, we have increased inference speed by 30x and reduced training memory by approximately 60%. © 2024 Association for Computational Linguistics.

关键词： Knowledge graph

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：