检索结果-内蒙古大学图书馆

Communication-Efficient Federated Learning With Compensated Overlap-FedAvg

IEEE TRANSACTIONS ON parallel AND distributed SYSTEMS 2022年第1期33卷 192-205页

作者： Zhou, Yuhao Ye, Qing Lv, Jiancheng Sichuan Univ Coll Comp Sci Chengdu 610017 Peoples R China

While petabytes of data are generated each day by a number of independent computing devices, only a few of them can be finally collected and used for deep learning (DL) due to the apprehension of data security and privacy leakage, thus seriously retarding the extension of DL. In such a circumstance, federated learning (FL) was proposed to perform model training by multiple clients' combined data without the dataset sharing within the cluster. Nevertheless, federated learning with periodic model averaging (FedAvg) introduced massive communication overhead as the synchronized data in each iteration is about the same size as the model, and thereby leading to a low communication efficiency. Consequently, variant proposals focusing on the communication rounds reduction and data compression were proposed to decrease the communication overhead of FL. In this article, we propose Overlap-FedAvg, an innovative framework that loosed the chain-like constraint of federated learning and paralleled the model training phase with the model communication phase (i.e., uploading local models and downloading the global model), so that the latter phase could be totally covered by the former phase. Compared to vanilla FedAvg, Overlap-FedAvg was further developed with a hierarchical computing strategy, a data compensation mechanism, and a nesterov accelerated gradients (NAG) algorithm. In Particular, Overlap-FedAvg is orthogonal to many other compression methods so that they could be applied together to maximize the utilization of the cluster. Besides, the theoretical analysis is provided to prove the convergence of the proposed framework. Extensive experiments conducting on both image classification and natural language processing tasks with multiple models and datasets also demonstrate that the proposed framework substantially reduced the communication overhead and boosted the federated learning process.

关键词： Collaborative work Data models Training Servers Computational modeling Convergence Deep learning distributed computing federated learning overlap efficient communication

来源：评论

学校读者我要写书评

暂无评论

Benchmarking Federated Learning on High-Performance Computing: Aggregation methods and Their Impact

Benchmarking Federated Learning on High-Performance Computin...

引用

Euromicro Conference on parallel, distributed and Network-Based processing

作者： Daniela Annunziata Marzia Canzaniello Martina Savoia Salvatore Cuomo Francesco Piccialli Department of Mathematics and Applications “R. Caccioppoli” University of Naples Federico II Naples Italy

Federated Learning (FL) diverges from traditional Machine Learning (ML) models decentralizing data utilization, addressing privacy concerns. This approach involves iterative model updates, where individual devices compute gradients based on local data, share updates with a central server, and receive an improved global model. High-Performance Computing (HPC) systems enhance FL efficiency by leveraging parallel processing. In this study, we aim to explore FL efficiency using four aggregation methods on three datasets across six clients, assess metrics like global model accuracy and communication efficiency, and evaluate FL on HPC. We employ Flower, a versatile FL framework, in our experiments. Our chosen datasets include MNIST, Digits, and Semeion Handwritten Digit, distributed among two clients each. We utilize NVIDIA GPUs for computation, with aggregation methods such as FedAvg, FedProx, FedOpt, and FedYogi. Metrics include Convergence Time, Global Model Accuracy, Communication Efficiency, and HPC Throughput. The results will provide insights into FL performance, especially in HPC environments, impacting convergence, communication, and resource utilization.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Detecting Adversarial Samples for Deep Learning Models: A Comparative Study

引用

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING 2022年第1期9卷 231-244页

作者： Zhang, Shigeng Chen, Shuxin Liu, Xuan Hua, Chengyao Wang, Weiping Chen, Kai Zhang, Jian Wang, Jianxin Cent South Univ Sch Comp Sci & Engn Changsha 410083 Hunan Peoples R China Chinese Acad Sci State Key Lab Informat Secur Inst Informat Engn Beijing 100093 Peoples R China Hunan Univ Coll Comp Sci & Elect Engn Changsha 410082 Hunan Peoples R China Sci & Technol Parallel & Distributed Proc Laborat Changsha 410073 Peoples R China Chinese Acad Sci Inst Informat Engn SKLOIS Beijing 100093 Peoples R China Univ Chinese Acad Sci Sch Cyber Secu Rity Beijing Peoples R China

Deep learning techniques such as convolutional neural networks (CNNs) have been used in a wide range of fields due to their superior performance, e.g., image classification, autonomous driving and natural language processing. However, recent progress shows that deep learning models are vulnerable to adversarial samples, which are crafted by adding small perturbations on normal samples that are imperceptible to human beings but can mislead the deep learning models to output incorrect results. Many adversarial attack models are proposed and many adversarial detection methods are developed to detect adversarial samples generated by these attack models. However, the evaluations of these detection methods are fragmented and scatter in separate literature, and the community still lacks a comprehensive understanding of the ability and performance of existing adversarial detection methods when facing different attack models on different datasets. In this paper, by using image classification as the example application scenario, we conduct a comprehensive study on the performance of five mainstream adversarial detection methods against five major attack models on four widely used benchmark datasets. We find that the detection accuracy of different methods interleaves for different attack models and dataset. Moreover, besides detection accuracy, we also evaluate the time efficiency of different detection methods. The findings reported in this paper can provide useful insights when designing systems to detect adversarial samples and act as a guideline to design new methods to detect adversarial samples.

关键词： Training Neural networks Deep learning Detectors Robustness Predictive models Safety Adversarial detection efficiency adversarial samples detection deep learning attacks image classification

来源：评论

学校读者我要写书评

暂无评论

引用

4th International Conference on Information Science, parallel and distributed Systems, ISPDS 2023

作者： Zhou, Ya Zhao, Ning Huang, Guimin Deng, Nanxiao Guo, Qingkai School of Computer Science and Information Security Guilin University of Electronic Technology Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin China

ISBN: (纸本)9798350337181

In sentence similarity research methods, sentence similarity is often calculated from semantic aspects, however, the influence of other features is ignored. For example, the influence of sentence syntactic structure and word order on sentence similarity is not considered. Therefore, This paper proposes a method to measure the degree of similarity between sentences based on a multi-feature fusion of semantics, syntactic similarity and word order similarity, respectively. For the aspect of semantic similarity, in this paper, we first preprocess the sentence and obtain the word vectors, and then obtains sentence vector features through neural networks, so as to calculate the sentence semantic similarity;Finally, for the aspect of word order similarity, word order is considered as the auxiliary feature information to calculate sentence similarity since word order has a smaller effect on sentence similarity compared to the above information. The experimental results show that the proposed model has an accuracy of 77.4 % and an F1 value of 84.3% on the MRPC dataset. © 2023 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Domain Adaptive Semantic Segmentation via image Translation and Representation Alignment 19

Domain Adaptive Semantic Segmentation via Image Translation ...

引用

19th IEEE International Symposium on parallel and distributed processing with Applications (IEEE ISPA)

作者： Kang, Jingxuan Zang, Bin Cao, Weipeng Univ Liverpool Sch Comp Sci Liverpool Merseyside England Beijing Inst Technol Sch Comp Sci & Technol Beijing Peoples R China Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen Peoples R China

ISBN: (纸本)9781665435741

Domain Adaptation for semantic segmentation is of vital significance since it enables effective knowledge transfer from a labeled source domain (i.e., synthetic data) to an unlabeled target domain (i.e., real images), where no effort is devoted to annotating target samples. Prior domain adaptation methods are mainly based on image-to-image translation model to minimize differences in image conditions between source and target domain. However, there is no guarantee that feature representations from different classes in the target domain can be well separated, resulting in poor discriminative representation. In this paper, we propose a unified learning pipeline, called image Translation and Representation Alignment (ITRA), for domain adaptation of segmentation. Specifically, it firstly aligns an image in the source domain with a reference image in the target domain using image style transfer technique (e.g., CycleGAN) and then a novel pixelcentroid triplet loss is designed to explicitly minimize the intraclass feature variance as well as maximize the inter-class feature margin. When the style transfer is finished by the former step, the latter one is easy to learn and further decreases the domain shift. Extensive experiments demonstrate that the proposed pipeline facilitates both image translation and representation alignment and significantly outperforms previous methods in both GTA5 -> Cityscapes and SYNTHIA -> Cityscapes scenarios.

关键词： Domain Adaptation Semantic Segmentation Style Transfer Centroid Alignment

来源：评论

学校读者我要写书评

暂无评论

A Fast Search Algorithm for SqueeSAR distributed Scatterers in the Problem of Calculating Displacement Velocities

引用

PROGRAMMING AND COMPUTER SOFTWARE 2021年第6期47卷 426-438页

作者： Popov, S. E. Potapov, V. P. Fed Res Ctr Informat & Computat Technol Pr Akad Lavrenteva 6 Novosibirsk 630090 Russia

This paper describes a software implementation of a fast distributed scatterer search algorithm for the problem of displacement velocity calculation based on the Apache Spark platform. A complete scheme for calculating displacement velocities by the persistent scatterer method is considered. The proposed algorithm is integrated into the scheme after the stage of subpixel-accuracy alignment of a stack of time-series images. The search for distributed scatterers is carried out independently in shift windows over the entire area of the image. The presence of distributed scatterers is determined based on the assumption that pairs of samples in the window, which are composed of vectors of complex pixel values in each of the N images, are homogeneous. This assumption stems from the fulfillment of the Kolmogorov-Smirnov criterion for each pair. Toestimate phases of homeogenic pixels, the maximization problem is solved. It is shown that the proposed algorithm is not iterative and can be implemented in the framework of the parallel computing paradigm. Toenable distributed in-memory processing of radar data arrays (from 60 images) across many physical nodes in a network environment, we use the Apache Spark parallel processing platform. In this case, the time it takes to find distributed scatterers is reduced by a factor of 10 on average as compared to a single-processor implementation of the algorithm. The comparative results of testing the computing system on a demo cluster are presented. The algorithm is implemented in Python with a detailed description of the objects and methods of the algorithm.

关键词： Computer Science general Software Engineering/Programming and Operating Systems Operating Systems Software Engineering Artificial Intelligence

来源：评论

学校读者我要写书评

暂无评论

A Counterfactual Ultrasound Anti-Interference Self-Supervised Network for B-mode Ultrasound Tongue Extraction

A Counterfactual Ultrasound Anti-Interference Self-Supervise...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal processing, ICASSP 2025

作者： Jia, Yan Cheng, Yuqing Xu, Kele Dou, Yong Qiao, Peng He, Zhouyu National Key Laboratory of Parallel and Distributed Computing College of Computer Science and Technology National University of Defense Technology Changsha China College of Systems Engineering National University of Defense Technology Changsha China

ISBN: (纸本)9798350368741

B-mode ultrasound tongue imaging is a non-invasive and real-time method for visualizing vocal tract deformation. However, accurately extracting the tongue's surface contour remains a significant challenge due to the low signal-to-noise ratio (SNR) and prevalent speckle noise in ultrasound images. Traditional supervised learning models often require large labeled datasets, which are labor-intensive to produce and susceptible to noise interference. To address these limitations, we present a novel Counterfactual Ultrasound Anti-Interference Self-Supervised Network (CUAI-SSN), which integrates self-supervised learning (SSL) with counterfactual data augmentation, progressively disentangles confounding factors, ensuring that the model generalizes well across varied ultrasound conditions. Our approach leverages causal reasoning to decouple noise from relevant features, enabling the model to learn robust representations that focus on essential tongue structures. By generating counterfactual image-label pairs, our method introduces alternative, noise-independent scenarios that enhance model training. Furthermore, we introduce attention mechanisms to enhance the network's ability to capture fine-grained details even in noisy conditions. Extensive experiments on real ultrasound tongue images demonstrate that CUAI-SSN outperforms existing methods, setting a new benchmark for automated contour extraction in ultrasound tongue imaging. Our code is publicly available at https://***/inexhaustible419/CounterfactualultrasoundAI. © 2025 IEEE.

关键词： contour extraction counterfactual inference self-supervised learning speech production ultrasound image

来源：评论

学校读者我要写书评

暂无评论

AMQGaussian: Efficient 3D Gaussian Representation with Asymmetric Mixed-precision Quantization

AMQGaussian: Efficient 3D Gaussian Representation with Asymm...

引用

International Symposium on parallel and distributed processing with Applications, ISPA

作者： Yong Kou Qing Ye Yanci Zhang Xia Yuan Jinlong He College of Computer Science Sichuan University Chengdu China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

3D Gaussian Splatting (3DGS) has recently gained increasing attention in novel-view scene synthesis. However, it requires millions of 3D Gaussian spheres to achieve high-quality rendered images, leading to substantial GPU resource demands for training and rendering. This paper has developed an efficient framework to address the challenges faced by 3DGS. 1) We propose an asymmetric mixed-precision quantization strategy to efficiently quantize and dequantize the parameters of 3D Gaussian Spheres (excluding position and opacity) and utilized the Straight Through Estimator method to address the issue of non-backpropagatable gradients post-quantization. This approach significantly reduces GPU memory usage and model storage space. 2) Using statistical methods, we identify optimal gradient thresholds to enhance the densification algorithm of 3D Gaussian spheres, thereby further improving performance. 3) To address the speed bottlenecks in parallel processing of low-precision data using CUDA, we introduce a fast atomic operation method, increasing training speed tenfold. Overall, we validate the effectiveness of our framework across various datasets. It reduces GPU usage by 50% during training, decreases GPU usage threefold during rendering, halves the model storage space and maintains image quality comparable to 3DGS.

关键词： Training image quality Solid modeling Three-dimensional displays Quantization (signal) Statistical analysis Graphics processing units parallel processing Rendering (computer graphics) image reconstruction

来源：评论

学校读者我要写书评

暂无评论

A Noise-Based Approach Augmented with Neural Cleanse and JPEG Compression to Counter Adversarial Attacks on image Classification Systems

A Noise-Based Approach Augmented with Neural Cleanse and JPE...

引用

Euromicro Conference on parallel, distributed and Network-Based processing

作者： Igor Kotenko Igor Saenko Oleg Lauta Nikita Vasiliev Vladimir Sadovnikov Laboratory of Computer Security Problems St. Petersburg Federal Research Center of the Russian Academy of Sciences (SPC RAS) Saint-Petersburg Russia Department of Telecommunications and Information Security Admiral Makarov State University of Maritime and Inland Shipping Saint-Petersburg Russia

ISBN: (数字)9798331524937

ISBN: (纸本)9798331524944

Adversarial attacks are now becoming quite a dangerous means of disrupting image processing systems that use machine learning methods for decision making. Therefore, developing effective countermeasures against adversarial attacks is becoming quite an important area of cybersecurity. The paper proposes a noise-based approach to countering adversarial attacks that is augmented with neural-cleanse and jpeg-compression technologies. The idea of the proposed approach is that adding noise distorts the effect of an adversarial attack, and neural cleaning and jpeg compression eliminate the consequences of such an effect. The paper examines the three most well-known types of adversarial attacks: Fast Gradient Sign Method, Zeroth Order Optimization and One Pixel Attack. These attacks manipulate input data, resulting in misclassification or incorrect predictions by exploiting high-frequency components that are undetectable to humans. The research was carried out on two datasets: MNIST-JPG and PC Parts images. Two types of noise were used: Gaussian and Poisson. During the experiments, optimal parameters for these types of noise were found, ensuring maximum accuracy of image recognition after exposure to adversarial attacks.

关键词： Resistance image recognition image coding image resolution Noise Transform coding Machine learning Distortion Optimization image classification

来源：评论

学校读者我要写书评

暂无评论

Towards a SOLAP Based Decision Support System for Forest Fire Management

Towards a SOLAP Based Decision Support System for Forest Fir...

引用

2023 International Conference on Earth Observation and Geo-Spatial Information, ICEOGI 2023

作者： Belaroussi, Asma Derbal, Khalissa Benabdellah, Rachid Belhadj-Aissa, Mostefa Image Processing and Radiation Laboratory Ltir Usthb Algiers Algeria Lsi Laboratory Usthb Algiers Algeria General Directorate of Forestry Dgf Algiers Algeria

ISBN: (纸本)9798350396720

Risk, suppression, and damage assessment are essential parts of forest fire management. The involved data is distributed and heterogeneous, concerning meteorology, topomorphology, and socio-economy. In addition, it has a multiscale spatiotemporal analysis aspect. These reinforce the need to use the appropriate methods and technologies to enhance decision-making. This research investigates the potential of adopting spatial data warehousing techniques to build a SOLAP-based DSS that supports the whole decisional chain. (i) source collection and integration;(ii) data repository design and population;(iii) SOLAP analytics;and finally, (iv) data visualization in numerous formats, including tabular, graphic, and dynamic maps. © 2023 IEEE.

关键词： Data visualization

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：