检索结果-内蒙古大学图书馆

24th ieee/ACM International symposium on Cluster, Cloud, and Internet Computing (CCGrid)

作者： Cernuda, Jaime Logan, Luke Gainaru, Ana Klasky, Scott Lofstead, Jay Kougkas, Anthony Sun, Xian-He IIT Chicago IL 60616 USA Oak Ridge Natl Lab Oak Ridge TN USA Sandia Natl Labs Albuquerque NM USA

ISBN: (纸本)9798350395679;9798350395662

Modern simulation workflows generate and analyze massive amounts of data using I/O libraries like Adios2 and NetCDF. Although extensive work has optimized the I/O processes during the simulation phase, executing analytical queries-which often require iterative traversals of large files for insights-is cumbersome and usually constrained by low I/O performance. Instead of waiting for the analysis phase to process queries, quantities can be derived asynchronously during data production and cached, speeding up future queries. In this work, we introduce a context-aware I/O layer named 'Hades.' It is designed to efficiently derive insights from selected quantities without compromising overall workflow performance. Hades actively and asynchronously computes and stores these quantities while the data is in transit. Hades leverages a hierarchical buffering system with data access-aware prefetching to ensure quick and timely access to relevant data. It offers a flexible query interface empowering users to easily define derived quantities and provide control over data placement decisions. Hades is implemented using an Adios2 plugin engine and the Hermes buffering platform, enabling transparent use by any Adios-powered application or workflow. Experimental results demonstrate performance improvements by up to 3-4x for tested real-world scientific producer-consumer workflows.

关键词： Active Storage Hierarchical Storage Context Awareness Metadata Management data Operator In-transit Computing

来源：评论

学校读者我要写书评

暂无评论

Class Incremental Learning for Light-Weighted Networks

引用

ieee TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第12期34卷 12210-12220页

作者： Tao, Zhe Yu, Lu Yao, Hantao Huang, Shucheng Xu, Changsheng Tianjin Univ Technol Sch Comp Sci & Engn Tianjin 300384 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Multimodal Artificial Intelligence S Beijing 100190 Peoples R China Jiangsu Univ Sci & Technol Coll Comp Sci Zhenjiang 212100 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China

Despite deep neural networks (DNNs) show impressive performance across diverse tasks, they suffer from catastrophic forgetting when dealing with continuous data streams. Incremental learning aims to alleviate this phenomenon and enable DNNs to accumulate new knowledge to cope with the ever-changing world. Recently numerous advanced methods have been developed to enhance the incremental learning capabilities of neural networks. However, these methods mainly focus on the large networks, neglecting the unique needs of edged-device applications, which is surprisingly under-investigated in previous literature. In this paper, we propose two strategies for transferring knowledge from large teacher networks to light-weighted networks in class incremental learning. Specifically, in cases where the initial task contains a large number of categories, our static teacher strategy involves transferring knowledge from the teacher to the student network on the initial task to enhance the plasticity of the student network, and applying regularization constraints on the subsequent task to improve its stability. In a more challenging scenario where each task includes an equal number of categories, the dynamic teacher strategy continuously guides the student network on each task. We evaluate the proposed methods on CIFAR100, Tiny-ImageNet and ImageNet-subset datasets with different types of light-weighted networks (MobileNet, ShuffleNet). We observed that effective knowledge transfer resulting in the student network achieving performance comparable or even outperform the teacher network. Extensive and detailed experiments conducted on three datasets demonstrated the simplicity and effectiveness of our proposed method. Comprehensive analysis are also conducted including different factors and visualization.

关键词： Task analysis Neural networks Knowledge engineering Incremental learning Training data models Streams light-weighted networks network distillation image classification

来源：评论

学校读者我要写书评

暂无评论

Interactive Visual analysis of Structure-borne Noise data

引用

ieee Transactions on visualization and Computer Graphics 2023年第1期29卷 778-787页

作者： Splechtna, Rainer Gracanin, Denis Todorovic, Goran Goja, Stanislav Bedic, Boris Hauser, Helwig Matkovic, Kresimir VRVis Research Center Austria Virginia Tech BlacksburgVA United States AVL-AST D.o.o. Zagreb Croatia University of Bergen Bergen Norway

Numerical simulation has become omnipresent in the automotive domain, posing new challenges such as high-dimensional parameter spaces and large as well as incomplete and multi-faceted data. In this design study, we show how interactive visual exploration and analysis of high-dimensional, spectral data from noise simulation can facilitate design improvements in the context of conflicting criteria. Here, we focus on structure-borne noise, i.e., noise from vibrating mechanical parts. Detecting problematic noise sources early in the design and production process is essential for reducing a product's development costs and its time to market. In a close collaboration of visualization and automotive engineering, we designed a new, interactive approach to quickly identify and analyze critical noise sources, also contributing to an improved understanding of the analyzed system. Several carefully designed, interactive linked views enable the exploration of noises, vibrations, and harshness at multiple levels of detail, both in the frequency and spatial domain. This enables swift and smooth changes of perspective;selections in the frequency domain are immediately reflected in the spatial domain, and vice versa. Noise sources are quickly identified and shown in the context of their neighborhood, both in the frequency and spatial domain. We propose a novel drill-down view, especially tailored to noise data analysis. Split boxplots and synchronized 3D geometry views support comparison tasks. With this solution, engineers iterate over design optimizations much faster, while maintaining a good overview at each iteration. We evaluated the new approach in the automotive industry, studying noise simulation data for an internal combustion engine. © 2022 ieee.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

A DEEP LEARNING-BASED MULTITASKING MODEL FOR HYPERSPECTRAL IMAGE analysis USING NOVEL TAIGA dataSET

A DEEP LEARNING-BASED MULTITASKING MODEL FOR HYPERSPECTRAL I...

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Chhapariya, Koushikey Benoit, Alexandre Buddhiraju, Krishna Mohan Kumar, Anil Indian Inst Technol Ctr Studies Resources Engn Mumbai Maharashtra India Univ Savoie Mt Blanc Annecy France ISRO Indian Inst Remote Sensing Dehra Dun Uttarakhand India

ISBN: (纸本)9798350360332;9798350360325

Hyperspectral imaging is essential for the detailed and accurate identification of materials and features across various applications, including environmental monitoring and agricultural assessment. However, processing large-scale hyperspectral images is time-consuming and requires huge computational resources attributed to their volume. Additionally, classification methods lacking spatial information and focusing only on spectral information are inadequate with the feature complexity of the dataset. Hence, to overcome the above-mentioned challenges, there is a dire need for advanced algorithms capable of efficiently processing large-scale hyperspectral data. This paper highlights the motivation and need for multitask learning models for hyperspectral datasets. A large-scale hyperspectral dataset called TAIGA, which comprises both categorical and continuous forest variables, is considered. The proposed deep learning model addresses data imbalance and loss function concerns by incorporating spectral as well as spatial information to develop a multitask model. This work emphasizes the importance of accounting for correlations between data and tasks when designing a relevant model. As a result, the study leads to more efficient model training, reduced data requirements, and potentially improved overall predictive capabilities. We achieved overall accuracy for categorical variables as high as 98.25% and mean absolute error as low as 0.019 for continuous variables.

关键词： Multitasking hyperspectral image CNN TAIGA large-scale hyperspectral dataset spectral-spatial classification

来源：评论

学校读者我要写书评

暂无评论

A Feature Extraction and analysis Method for Battery Health Monitoring 33

A Feature Extraction and Analysis Method for Battery Health ...

引用

33rd International symposium on Industrial Electronics (ISIE)

作者： Tian, Jilun Zhang, Jiusi Luo, Hao Huang, Congsheng Chow, Mo-yuen Jiang, Yuchen Yin, Shen Harbin Inst Technol Dept Control Sci & Engn Harbin 150001 Peoples R China Natl Taipei Univ Technol Dept Elect Engn Taipei 10608 Taiwan North Carolina State Univ Dept Elect & Comp Engn Raleigh NC 27606 USA Norwegian Univ Sci & Technol Fac Engn N-7034 Trondheim Norway

ISBN: (纸本)9798350394085;9798350394092

Batteries are widely applied in industrial systems, promoting industrial production efficiency. The system prognosis needs to establish a battery health monitoring model and maintain regular maintenance, improving reliability and safety while reducing unnecessary economic losses. An accurate data-driven state-of-health (SOH) estimation method is crucial, as it mainly relies on the information related to battery SOH values in the extracted features. Therefore, we propose a statistical-based feature extraction method, which makes it physically meaningful and effective. To explore more information, we propose a feature analysis framework to analyze battery features, including visualization, correlation, and importance analysis. We utilize a random forest regression model as an effective learner and validate it based on NASA's real-world dataset. The feature analysis contents are beneficial for understanding the role and importance of features in SOH estimation tasks, which can also provide an extended scenario for feature extraction of batteries.

关键词： Battery health monitoring feature extraction state-of-health estimation random forest

来源：评论

学校读者我要写书评

暂无评论

TGVE: a tool for analysis and visualization of geospatial data

TGVE: a tool for analysis and visualization of geospatial da...

引用

25th Eurographics Conference on visualization, EuroVis 2023

作者： Hama, L. Beecham, R. Lomax, N. Leeds Institute for Data Analytics University of Leeds United Kingdom School of Geography University of Leeds United Kingdom

ISBN: (纸本)9783038682196

We introduce the Turing Geovisualisation Engine (TGVE), a web-based, open-source tool for interactive visualization and analysis of geospatial data. Built on ReactJS and R, TGVE is designed to support a variety of users, including data scientists and stakeholders who wish to engage the wider public with geospatial data. In this short paper, we provide an overview of TGVE’s features and capabilities, including its ability to publish data and customize visualization settings using URL parameters. We highlight the potential impact of TGVE for geospatial research and offer examples of its use in practice. Additionally, we discuss current limitations of the tool and outline future work, such as improving compatibility with other geospatial data formats and addressing performance issues for large datasets. © 2023 The Authors.

关键词： visualization

来源：评论

学校读者我要写书评

暂无评论

ChatVis: Automating Scientific visualization with a large Language Model

ChatVis: Automating Scientific Visualization with a Large La...

引用

2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and analysis, SC Workshops 2024

作者： Mallick, Tanwi Yildiz, Orcun Lenz, David Peterka, Tom Argonne National Laboratory LemontIL United States

ISBN: (纸本)9798350355543

We develop an iterative assistant we call "ChatVis"that can synthetically generate Python scripts for data analysis and visualization using a large language model (LLM). The assistant allows a user to specify the operations in natural language, attempting to generate a Python script for the desired operations, prompting the LLM to revise the script as needed until it executes correctly. The iterations include an error detection and correction mechanism that extracts error messages from the execution of the script and subsequently prompts LLM to correct the error. Our method demonstrates correct execution on five canonical visualization scenarios, comparing results with ground truth. We also compared our results with scripts generated by several other LLMs without any assistance. In every instance, ChatVis successfully generated the correct script, whereas the unassisted LLMs failed to do so. The code is available on GitHub: https://***/tanwimallick/ChatVis/. © 2024 ieee.

关键词： coding assistant large language models ParaView Scientific visualization synthetic script generation

来源：评论

学校读者我要写书评

暂无评论

Towards a Visual Perception-Based analysis of Clustering Quality Metrics

Towards a Visual Perception-Based Analysis of Clustering Qua...

引用

2024 Conference on visualization in data Science

作者： Blasilli, Graziano Kerrigan, Daniel Bertini, Enrico Santucci, Giuseppe Sapienza Univ Roma Rome Italy Northeastern Univ Boston MA USA

ISBN: (纸本)9798331528423;9798331528430

Clustering is an essential technique across various domains, such as data science, machine learning, and explainable artificial intelligence. Information visualization and visual analytics techniques have been proven to effectively support human involvement in the visual exploration of clustered data to enhance the understanding and refinement of cluster assignments. To support the human involvement, several perceptual studies and visual quality metrics have already been proposed. However, the visual perception of clustering quality metrics, also known as Cluster Validity Indexes (CVIs), still remains to be further explored. This paper presents the first attempt of a deep and exhaustive evaluation of the perceptive aspects of clustering quality metrics, focusing on the Davies-Bouldin Index, Dunn Index, Calinski-Harabasz Index, and Silhouette Score. Our research is centered around two main objectives: a) assessing the human perception of common CVIs in 2D scatterplots and b) exploring the potential of large Multimodal Models, in particular GPT-4o, to emulate the assessed human perception. To this end, we conducted two systematic data studies and a user study covering a broad collection of datasets. By discussing the obtained results, highlighting limitations, and areas for further exploration, this paper aims to propose a foundation for future research activities.

关键词： Clustering Clustering Quality Metrics Clustering Validity Indexes Visual Perception large Multimodal Models LMM

来源：评论

学校读者我要写书评

暂无评论

Efficient Interactive Global Cellular Signal Strength visualization

引用

ieee TRANSACTIONS ON BIG data 2022年第5期8卷 1209-1219页

作者： Ma, Mengyu Chen, Luo Ouyang, Xue Liu, Xiaoran Li, Jun Jing, Ning Natl Univ Def Technol Coll Elect Sci & Technol Changsha 410073 Peoples R China

Cellular Signal Strength (CSS), defined as the signal power received by mobile phones, is an important aspect of geographic information flow analysis, because the density of such information can reflect the urbanization variables such as population, gross domestic product, built-up area, electric power consumption, etc. Despite the importance, the real-time analysis of global CSS distribution remains a challenging problem due to the large data scale. In this article, a Display-driven Computing (DisDC) technique is designed and applied to provide efficient large scale interactive CSS visualization, generating results by calculating the value of each pixel that directly for display. Specifically, we present an efficient CSS measurement algorithm, which introduces spatial indexes and a corresponding query strategy;besides, an optimized parallel computing architecture is proposed to ensure the ability of real-time visualization. Experiments show that our approach obviously outperforms traditional methods and is capable of handling more than 40 million base stations in real-time. Moreover, an online demonstration is provided at https://***/MemoryMmy/CSSMap.

关键词： Cascading style sheets data visualization Base stations Real-time systems Big data Spatial databases visualization Cellular signal strength spatial analysis real-time big data parallel computing

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Potential of SDGSAT-1 Glimmer Imagery for Urban Road Detection

引用

ieee JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2025年 18卷 785-794页

作者： Wang, Yu Huang, Hailan Wu, Bin Sun Yat Sen Univ Sch Geospatial Engn & Sci Zhuhai 519082 Peoples R China

Sustainable Development Science Satellite 1 (SDGSAT-1), an innovative satellite engineered by the Chinese Academy of Sciences, focuses on achieving sustainable development goals (SDGs) by using high-resolution low-light imaging for urban planning and environmental monitoring. Although its potential to provide substantial information relevant to various SDG indicators has been demonstrated across multiple domains, a comprehensive analysis of the potential of SDGSAT-1 glimmer imagery for urban road detection has not yet conducted. To address this gap, we aimed to evaluate the applicability of SDGSAT-1 glimmer imagery in road detection across five cities with well-developed road networks in China, namely Chengdu, Guangzhou, Hangzhou, Shanghai, and Wuhan. Watershed segmentation and optimal thresholding algorithms were applied to extract the road network data of these five cities. We observed that the large connecting roads, including highways (Motorway), urban expressways (Trunk), urban main roads (Primary), urban secondary roads (Secondary), and urban tertiary roads (Tertiary), were visible in the SDGSAT-1 glimmer data. Watershed segmentation and optimal thresholding algorithms demonstrated a remarkable precision of approximately 80% for road extraction across all cities. Approximately, 64% of urban roads were detectable within the SDGSAT-1 glimmer imager data. Our study suggests that the SDGSAT-1 glimmer data have great potential for extracting urban roads.

关键词： Roads Watersheds Urban areas data mining Image segmentation Transportation Sustainable development Economics visualization Thresholding (Imaging) Glimmer data nighttime light (NTL) sustainable development science satellite 1 (SDGSAT-1) urban roads watershed segmentation

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：