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： Yang, Siling He, Shuibing Wang, Wenjiong Yin, Yanlong Wu, Tong Chen, Weijian Zhang, Xuechen Sun, Xian-He Feng, Dan The State Key Laboratory of Blockchain and Data Security Zhejiang University China Zhejiang Lab China Institute of Blockchain and Data Security China Zhejiang Key Laboratory of Big Data Intelligent Computing China Washington State University Vancouver United States Illinois Institute of Technology United States Huazhong University of Science and Technology China Wuhan National Laboratory for Optoelectronics China

ISBN: (纸本)9798331506476

Graph convolutional networks (GCNs) are popular for a variety of graph learning tasks. ReRAM-based processing-in-memory (PIM) accelerators are promising to expedite GCN training owing to their in-situ computing capability. However, existing accelerators can be severely underutilized even with pipelines, due to the oversight of the skewed execution times of various GCN stages and the ignorance of skewed degrees of graph vertices. In this work, we propose GOPIM, a GCN-oriented pipeline optimization for PIM accelerators to expedite GCN training. First, GOPIM proposes an ML-based scheme that allocates crossbar resources to the most needed stages to streamline the overall pipeline. Second, GOPIM utilizes a selective vertex updating technique that evenly distributes vertices on crossbars by interleaved mapping. These techniques collectively reduce the overall execution time without losing much accuracy. We also provide a practical architecture design for GOPIM. Our experimental results show that, GoPIM achieves up to 191 × speedup and 16.1 × energy saving, compared to the state-of-the-art work. © 2025 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

MDPFL:A Multiple Differential Privacy Protection Method based on Federated Learning 26

MDPFL:A Multiple Differential Privacy Protection Method base...

引用

26th International Conference on Computer Supported Cooperative Work in Design, CSCWD 2023

作者： Liu, Xinyu Xu, Xiaolong Nanjing University of Posts and Telecommunications Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing China

ISBN: (纸本)9798350331684

With the continuous development of artificial intelligence technology, machine learning in distributed network systems, such as IoVflntemet of Vehicles), will inevitably lead to privacy leakage. At present, there are lots of problems in federated learning, differential privacy and other machine learning privacy protection schemes, such as high loss of availability and the need of a fully trusted third-party server. In order to solve these problems, we propose MDPFL(Multiple Differential Privacy based on Federated Learning) algorithm. The algorithm combines with the differential privacy model in each stage of federated learning to solve the problem of curious third-party data collectors obtaining users' original data in the process of machine learning. Meanwhile, the algorithm does not adopt a decentralized machine learning scheme directly, but uses a double noise adding mode with the existence of the third-party data collectors. We use Laplacian mechanism in the central server and GRR mechanism in local clients to ensure the goal of stability of the effect of the machine ¡earning model. The algorithm is compared with FedAvg, CDP, LDP algorithm in accuracy and loss rate which based on EMNIST data sets. In different global sensitivities, the model training effect is consistent with FedAvg algorithm while comparing with LDP algorithm, the speed of model convergence is improved. © 2023 IEEE.

关键词： Learning algorithms

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

Complementary Label Queries for Efficient Active Learning 23

Complementary Label Queries for Efficient Active Learning

引用

6th International Conference on Image and Graphics Processing, ICIGP 2023

作者： Liu, Shengyuan Hu, Tianlei Chen, Ke Mao, Yunqing Key Lab of Intelligent Computing Based Big Data of Zhejiang Province Zhejiang University China Co. Ltd. China

ISBN: (纸本)9781450398572

Many active learning methods are based on the assumption that a learner simply asks for the true labels of some training data from annotators. Unfortunately, it is expensive to exactly annotate instances in real-world classification tasks. To mitigate this problem, we propose a novel active learning setting, named active learning with complementary labels (ALCL). The ALCL learners only ask yes/no questions in some classes. After receiving answers from annotators, ALCL learners get a few supervised data and more training instances with complementary labels, which only specify one of the classes to which the pattern does not belong. There are two challenging issues in ALCL: the first is how to learn from these complementary labels, and another is how to sample instances to be queried. For the first issue, we propose a novel method that redistributes the weights of instances based on the balance of category contribution to learn from ordinary labels and complementary labels. For the second issue, we propose a weighting mechanism to improve existing uncertainty-based sampling strategies under this novel setup. Experiments on various datasets demonstrate the feasibility of ALCL and the superiority of our approaches. © 2023 ACM.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Image Matching Algorithm Under Adverse Conditions Based on SIFT 14

Image Matching Algorithm Under Adverse Conditions Based on S...

引用

14th International Conference on Information Technology in Medicine and Education, ITME 2024

作者： Chen, Hao Cao, Xinrong Li, Zuoyong Lin, Lihui Fuzhou University College of Computer and Big Data Fuzhou350108 China Minjiang University Fujian Provincial Key Laboratory of Information Processing and Intelligent Control School of Computer and Big Data Fuzhou350121 China Wuyi University Fujian Key Laboratory of Big Data Application and Intellectualization for Tea Industry School of Mathematics and Computer Science Fujian354300 China

ISBN: (纸本)9798350356670

Image matching technology is crucial in computer vision applications. However, the traditional SIFT (Scale-Invariant Feature Transform) algorithm often faces challenges under adverse conditions, such as a high number of mismatched points and few accurately matched points. Additionally, in practical scenarios, environmental factors like lighting changes and image blurring further reduce image similarity. Therefore, this paper proposes an improved SIFT-based image matching algorithm to address these challenges. First, SIFT is used to extract image features, and a bidirectional FLANN (Fast Library for Approximate Nearest Neighbors) matching strategy is employed to initially filter out incorrect features. Then, the RANSAC (Random Sample Consensus) algorithm is used to screen out unreliable matching point pairs. Finally, the matching accuracy is evaluated based on the remaining pure matching point pairs. Experimental results demonstrate that, compared to the traditional SIFT algorithm, this improved algorithm achieves higher matching accuracy and robustness under adverse conditions, exhibiting superior performance. © 2024 IEEE.

关键词： Image matching

来源：评论

学校读者我要写书评

暂无评论

Distributed Truss Computation in Dynamic Graphs

引用

Tsinghua Science and Technology 2023年第5期28卷 873-887页

作者： Ziwei Mo Qi Luo Dongxiao Yu Hao Sheng Jiguo Yu Xiuzhen Cheng School of Computer Science and Technology Shandong UniversityQingdao 266200China State Key Laboratory of Software Development Environment School of Computer Science and Engineering and the Beijing Advanced Innovation Center for Big Data and Brain ComputingBeihang UniversityBeijing 100191China Big Data Institute Qilu University of Technology(Shandong Academy of Sciences)Jinan 250353China

Large-scale graphs usually exhibit global sparsity with local cohesiveness,and mining the representative cohesive subgraphs is a fundamental problem in graph *** k-truss is one of the most commonly studied cohesive subgraphs,in which each edge is formed in at least k 2 triangles.A critical issue in mining a k-truss lies in the computation of the trussness of each edge,which is the maximum value of k that an edge can be in a *** works mostly focus on truss computation in static graphs by sequential ***,the graphs are constantly changing dynamically in the real *** study distributed truss computation in dynamic graphs in this *** particular,we compute the trussness of edges based on the local nature of the k-truss in a synchronized node-centric distributed *** decomposing the trussness of edges by relying only on local topological information is possible with the proposed distributed decomposition ***,the distributed maintenance algorithm only needs to update a small amount of dynamic information to complete the *** experiments have been conducted to show the scalability and efficiency of the proposed algorithm.

关键词： distributed algorithm dynamic graph graph mining cohesive subgraph k-truss

来源：评论

学校读者我要写书评

暂无评论

Research progress of quantum artificial intelligence in smart city

引用

intelligent and Converged Networks 2024年第2期5卷 116-133页

作者： Sumin Wang Ning Wang Tongyu Ji Yiyun Shi Chao Wang the Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication Shanghai UniversityShanghai 200444China School of Information Engineering Ganzhou Key Laboratory of Cloud Computing and Big Data ResearchGannan University of Science and TechnologyGanzhou 341000China

The rapid accumulation of big data in the Internet era has gradually decelerated the progress of Artificial Intelligence(AI).As Moore’s Law approaches its limit,it is imperative to break the constraints that are holding back artificial *** computing and artificial intelligence have been advancing along the highway of human civilization for many years,emerging as new engines driving economic and social *** article delves into the integration of quantum computing and artificial intelligence in both research and *** introduces the capabilities of both universal quantum computers and special-purpose quantum computers that leverage quantum *** discussion further explores how quantum computing enhances classical artificial intelligence from four perspectives:quantum supervised learning,quantum unsupervised learning,quantum reinforcement learning,and quantum deep *** an effort to address the limitations of smart cities,this article explores the formidable potential of quantum artificial intelligence in the realm of smart *** does so by examining aspects such as intelligent transportation,urban operation assurance,urban planning,and information communication,showcasing a plethora of practical achievements in the *** the foreseeable future,Quantum Artificial Intelligence(QAI)is poised to bring about revolutionary development to smart *** urgency lies in developing quantum artificial intelligence algorithms that are compatible with quantum computers,constructing an efficient,stable,and adaptive hybrid computing architecture that integrates quantum and classical computing,preparing quantum data as needed,and advancing controllable qubit hardware equipment to meet actual *** ultimate goal is to shape the next generation of artificial intelligence that possesses common sense cognitive abilities,robustness,excellent generalization capabilities,and interpretability.

关键词： quantum artificial intelligence smart city D-Wave quantum computer

来源：评论

学校读者我要写书评

暂无评论

Privacy protection oriented on data stream computing

Privacy protection oriented on data stream computing

引用

2022 International Conference on Computer Application and Information Security, ICCAIS 2022

作者： He, Renjie Zheng, Xianghan Wang, Gong School of Computer and Big Data Fuzhou University Fujian Fuzhou China The Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fujian Fuzhou China

ISBN: (纸本)9781510663459

data security has always been one of the public concerns, and homomorphic encryption technology, as one of the effective means to ensure data security, has always been a hot topic of academic research. Traditional machine learning is trained in a batch environment. In practice, data owners tend to outsource data and model training tasks to third-party clouds with huge computing resources, but data outsourcing carries the risk of privacy leakage. Therefore, in order to prevent this situation, an effective method is to encrypt the data before the outsourcing, and the third-party cloud trains the machine model of the ciphertext data. This method has special requirements for encryption algorithms which should support computation directly on the ciphertext data. This paper presents a novel data flow computing privacy protection framework based on the homomorphic encryption algorithm CKKS and the flow data processing engine Flink, called SDPPF (Streaming data Privacy Protection on Flink). The proposed framework supports the functions of data stream encryption, decryption and ciphertext computing, and expands the functions of vector point multiplication and array sum operation on the basic simultaneous operation supported by the original CKKS algorithm. This paper also selects a classical machine learning algorithm: KNN algorithm, combined with the homomorphic encryption algorithm CKKS to realize the privacy protection machine learning algorithm. © 2023 SPIE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

MS-SAM: Multi-scale SAM Based on Dynamic Weighted Agent Attention 31st

MS-SAM: Multi-scale SAM Based on Dynamic Weighted Agent Att...

引用

31st International Conference on Multimedia Modeling, MMM 2025

作者： Yang, Enhui Zhang, Zhibin Engineering Research Center of Ecological Big Data Ministry of Education Key Laboratory of Wireless Networks and Mobile Computing Inner Mongolia University Hohhot010021 China

ISBN: (纸本)9789819620630

The Segment Anything Model (SAM), introduced in 2023, has made significant advancements in the field of computer vision. However, SAM faces two major challenges: limitations related to single-scale processing and high computational demands for large-scale data handling, which restrict its widespread adoption in practical applications. Therefore, we propose MS-SAM to achieve better segmentation performance at a lower cost. We introduce a multi-scale image encoder to handle images of varying scales and complexities. Additionally, we propose a dynamic weighted agent attention, which can flexibly and efficiently adjust attention distribution while reducing computational costs. We train MS-SAM on a 1/20 subset of the SA-1B dataset for instance segmentation using points or boxes as prompts. The results show that we maintain the accuracy with the SAM method while reducing the parameters by about 6 times. We give sufficient experimental results to demonstrate its effectiveness. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

A method of multi-models fusion for speaker recognition

引用

International Journal of Speech Technology 2022年第2期25卷 493-498页

作者： Wu, Hao Luo, Linkai Peng, Hong Wen, Wei Department of Automation Xiamen University Xiamen China Xiamen Key Laboratory of Big Data Intelligent Analysis and Decision Xiamen China

As a new type of biometrics recognition technology, speaker recognition is gaining more and more attention because of the advantages in remote authentication. In this paper, we construct an end-to-end speaker recognition model named GAPCNN in which a convolutional neural network is used to extract speaker embeddings from spectrogram, and speaker recognition is performed by the cosine similarity of embeddings. In addition, we use global average pooling instead of the traditional temporal average pooling to adapt to different voice lengths. We use the ‘dev’ set of Voxceleb2 for training, then evaluate the model in the test set of Voxceleb1, and obtain an equal error rate (EER) of 4.04%. Furthermore, we fuse our GAPCNN with the x-vector model and the thin-Resnet model with GhostVLAD, and obtain an EER of 3.01% which is better than any of the three. It indicates that GAPCNN is an important complement to the x-vector model and the thin-Resnet model with GhostVLAD. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

关键词： Embeddings

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：