检索结果-内蒙古大学图书馆

A Synchronous Parallel Method with Parameters Communication Prediction for Distributed Machine Learning 19th

A Synchronous Parallel Method with Parameters Communication...

19th EAI International Conference on Collaborative Computing: Networking, Applications and Worksharing, CollaborateCom 2023

作者： Zeng, Yanguo Xue, Meiting Xu, Peiran Shi, Yukun Zeng, Kaisheng Zhang, Jilin Yue, Lupeng School of Computer Science and Technology Hangzhou Dianzi University Hangzhou310018 China Key Laboratory for Modeling and Simulation of Complex Systems Ministry of Education Hangzhou310018 China Data Security Governance Zhejiang Engineering Research Center Hangzhou310018 China National University of Defense Technology Changsha China School of Cyberspace HangZhou Dianzi University Hangzhou China Department of Computer Science and Technology Tsinghua University Beijing China

ISBN: (纸本)9783031545306

With the development of machine learning technology in various fields, such as medical care, smart manufacturing, etc., the data has exploded. It is a challenge to train a deep learning model for different application domains with large-scale data and limited resources of a single device. The distributed machine-learning technology, which uses a parameter server and multiple clients to train a model collaboratively, is an excellent method to solve this problem. However, it needs much communication between different devices with limited communication resources. The stale synchronous parallel method is a mainstream communication method to solve this problem, but it always leads to high synchronization delay and low computing efficiency as the inappropriate delay threshold value set by the user based on experience. This paper proposes a synchronous parallel method with parameters communication prediction for distributed machine learning. It predicts the optimal timing for synchronization, which can solve the problem of long synchronization waiting time caused by the inappropriate threshold settings in the stale synchronous parallel method. Moreover, it allows fast nodes to continue local training while performing global synchronization, which can improve the resource utilization of work nodes. Experimental results show that compared with the delayed synchronous parallel method, the training time and quality, and resource usage of our method are both significantly improved. © ICST Institute for Computer sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Enhanced carbon capture with motif-rich amino acid loaded defective robust metal-organic frameworks

引用

Nano Research 2024年第3期17卷 2004-2010页

作者： Qi-Ye Ju Jia-Jia Zheng Li Xu Hai-Yan Jiang Zi-Qian Xue Lu Bai Yang-Yang Guo Ming-Shui Yao Ting-Yu Zhu State Key Laboratory of Multi-phase Complex Systems CAS Key Laboratory of Green Process and EngineeringBeijing Key Laboratory of Ionic Liquids Clean ProcessInstitute of Process EngineeringChinese Academy of SciencesBeijing 100190China Laboratory of Theoretical and Computational Nanoscience National Center for Nanoscience and TechnologyChinese Academy of SciencesBeijing 100190China School of Rare Earths University of Science and Technology of ChinaHefei 230026China Ganjiang Innovation Academy Chinese Academy of SciencesGanzhou 341000China University of Chinese Academy of Sciences Chinese Academy of SciencesBeijing 100049China Institute for Integrated Cell-Material Sciences Kyoto UniversityInstitute for Advanced StudyKyoto UniversityYoshidaUshinomiya-choSakyokuKyoto 606-8501Japan

The use of metal-organic frameworks(MOFs)as solid adsorption materials for carbon capture is promising,but achieving efficient and reversible adsorption with a balance of capacity and selectivity for carbon dioxide(CO_(2))over N_(2) remains a *** take full advantage of the strong channel traffic and robustness of MOFs with relatively small pores,it is highly necessary to employ a defect-engineering strategy to construct a broader channel structure that can facilitate the loading of functional motif-rich amino acids(AAs).This strategy can greatly enhance the CO_(2) adsorption performance of *** this study,motif-rich amino acids are loaded into the defective and robust porous frameworks via combined defect-engineering and post-synthetic *** defective Zr/Hf-MOF-808s modified with AAs,especially for the 18 mol%4-nitroisophthalic acid,generated defective products allowing for the loading of L-serine(L-Ser).This modification resulted in a significant improvement in both the adsorption capacity(248%improvement at 298 K,100 kPa)and the selectivity of CO_(2)/N_(2) using the ideal adsorbed solution theory(IAST),with the selectivity increasing to 120.55 and 38.27 at 15 and 100 kPa,respectively,while maintaining good cycling *** functional theory(DFT)simulation,CO_(2) temperature-programmed desorption(CO_(2)-TPD),and in situ Fourier transform infrared spectroscopy(FTIR)were further employed to have a better understanding of the enhanced CO_(2) adsorption ***,unlike the AAs loaded pristine MOF-808s that showed the best CO_(2) adsorption capacity with the loading of short and small glycine(Gly),the broadened channel size in our work enables the loading of functional motif-rich L-serine,which brings more active binding sites,improving CO_(2) adsorption.

关键词： carbon capture and storage metal-organic framework(MOF) amino acids(AAs) defect-engineering motif-rich

来源：评论

学校读者我要写书评

暂无评论

Exploring the Physical World Adversarial Robustness of Vehicle Detection

arXiv

引用

arXiv 2023年

作者： Jiang, Wei Zhang, Tianyuan Liu, Shuangcheng Ji, Weiyu Zhang, Zichao Xiao, Gang Information Science Academy China Eletronics Technology Group Corporation Beijing100846 China Beihang University Beijing China National Key Laboratory for Complex Systems Simulation Beijing China

Adversarial attacks can compromise the robustness of real-world detection models. However, evaluating these models under real-world conditions poses challenges due to resource-intensive experiments. Virtual simulations offer an alternative, but the absence of standardized benchmarks hampers progress. Addressing this, we propose an innovative instant-level data generation pipeline using the CARLA simulator. Through this pipeline, we establish the Discrete and Continuous Instant-level (DCI) dataset, enabling comprehensive experiments involving three detection models and three physical adversarial attacks. Our findings highlight diverse model performances under adversarial conditions. Yolo v6 demonstrates remarkable resilience, experiencing just a marginal 6.59% average drop in average precision (AP). In contrast, the ASA attack yields a substantial 14.51% average AP reduction, twice the effect of other algorithms. We also note that static scenes yield higher recognition AP values, and outcomes remain relatively consistent across varying weather conditions. Intriguingly, our study suggests that advancements in adversarial attack algorithms may be approaching its "limitation".In summary, our work underscores the significance of adversarial attacks in real-world contexts and introduces the DCI dataset as a versatile benchmark. Our findings provide valuable insights for enhancing the robustness of detection models and offer guidance for future research endeavors in the realm of adversarial attacks. Copyright © 2023, The Authors. All rights reserved.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Thermal convection driven by a heat-releasing scalar component

引用

Acta Mechanica Sinica 2022年第10期38卷 56-65页

作者： Yuhang Du Mengqi Zhang Yantao Yang State Key Laboratory for Turbulence and Complex Systems Department of Mechanics and Engineering ScienceCollege of Engineeringand Beijing Innovation Center for Engineering Science and Advanced TechnologyPeking UniversityBeijing100871China Department of Mechanical Engineering National University of SingaporeSingapore117575Singapore

In this paper,we investigate the thermal convection flow which is driven by a heat-releasing concentration *** from our previous work on the internally heated double diffusive convection(IHDDC),in the current internally heated Rayleigh-Bénard convection(IHRBC),the fluid density depends solely on the temperature field and the concentration field only serves as the internal heat *** stability analyses reveal that the most unstable mode is always the stationary *** critical Rayleigh number,which measures the strength of the unstable driving force,decreases with the Schmidt number(the ratio between the viscosity and the molecular diffusivity of concentration field),but increases with the Prandtl number(the ratio between the viscosity and the thermal molecular diffusivity).Fully developed flows are then studied by three-dimensional direct numerical *** unifying model developed for IHDDC can also be used to describe the transport properties for the current *** characteristic widths are smaller for the plumes descending from the top plate than those ascending from the bottom one.

关键词： Internally heated convection Direct numerical simulation

来源：评论

学校读者我要写书评

暂无评论

FedAG:A Federated Learning Method Based on Data Importance Weighted Aggregation

FedAG:A Federated Learning Method Based on Data Importance W...

引用

2023 IEEE/CIC International Conference on Communications in China, ICCC 2023

作者： Xu, Mengchu Zeng, Yan Xue, Meiting Zhang, Ji-Lin Wan, Jian Zhou, Mingyao Wen, Yilin Shi, Yukun Hangzhou Dianzi University School of Computer Science and Technology Hangzhou China Key Laboratory of Complex Systems Modeling and Simulation Ministry of Education Hangzhou China Zhejiang Engineering Research Center of Data Security Governance Hangzhou China Hangzhou Dianzi University School of Cyberspace Hangzhou China Huawei Enterprise Communication Technology Co Hangzhou China

ISBN: (纸本)9798350345384

Federated learning is a distributed machine learning approach that trains models with multiple clients and data locally. However, the existing methods ignore the differences between local models caused by the data heterogeneity, and give a high aggregation weight for the low-quality local model, which leads to low accuracy of the global model. For this problem, we propose a federated learning method based on data importance weighted aggregation, FedAG. It constructs an evaluation model of data importance based on the data distribution skewness, data volume, and data quality of clients, and aggregates local models with different weights computed by the evaluation model of data importance into a global model. We make experiments on five datasets including EMNIST, MNIST, and CIFAR-10. The results show that the accuracy of FedAG is respectively improved by 6.5%, 1.8%, 1.3% and 2.1%, compared with FedAvg, FedProx, FedVar, and Scaffold algorithms. © 2023 IEEE.

关键词： Scaffolds

来源：评论

学校读者我要写书评

暂无评论

Prompting Large Language Models with Answer Heuristics for Knowledge-Based Visual Question Answering

Prompting Large Language Models with Answer Heuristics for K...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zhenwei Shao Zhou Yu Meng Wang Jun Yu Key Laboratory of Complex Systems Modeling and Simulation School of Computer Science and Technology Hangzhou Dianzi University China School of Computer Science and Information Engineering Hefei University of Technology China

Knowledge-based visual question answering (VQA) requires external knowledge beyond the image to answer the question. Early studies retrieve required knowledge from explicit knowledge bases (KBs), which often introduces irrelevant information to the question, hence restricting the performance of their models. Recent works have sought to use a large language model (i.e., GPT-3 [3]) as an implicit knowledge engine to acquire the necessary knowledge for answering. Despite the encouraging results achieved by these methods, we argue that they have not fully activated the capacity of GPT-3 as the provided input information is insufficient. In this paper, we present Prophet-a conceptually simple framework designed to $prompt$ GPT-3 with answer heuristics for knowledge-based VQA. Specifically, we first train a vanilla VQA model on a specific knowledge-based VQA dataset without external knowledge. After that, we extract two types of complementary answer heuristics from the model: answer candidates and answer-aware examples. Finally, the two types of answer heuristics are encoded into the prompts to enable GPT-3 to better comprehend the task thus enhancing its capacity. Prophet significantly outperforms all existing state-of-the-art methods on two challenging knowledge-based VQA datasets, OK-VQA and A-OKVQA, delivering 61.1% and 55.7% accuracies on their testing sets, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Continuous and comprehensive atmospheric observations in Beijing:a station to understand the complex urban atmospheric environment

引用

Big Earth Data 2020年第3期4卷 295-321页

作者： Yongchun Liu Chao Yan Zemin Feng Feixue Zheng Xiaolong Fan Yusheng Zhang Chang Li Ying Zhou Zhuohui Lin Yishou Guo Ying Zhang Li Ma Wenshuo Zhou Zhikun Liu Lubna Dada Kaspar Dallenbach Jenni Kontkanen Runlong Cai Tommy Chan Biwu Chu Wei Du Lei Yao Yonghong Wang Jing Cai Juha Kangasluoma Tom Kokkonen Joni Kujansuu Anton Rusanen Chenjuan Deng Yueyun Fu Rujing Yin Xiaoxiao Li Yiqun Lu Yiliang Liu Chaofan Lian Dongsen Yang Weigang Wang Maofa Ge Yuesi Wang Douglas R.Worsnop Heikki Junninen Hong He Veli-Matti Kerminen Jun Zheng Lin Wang Jingkun Jiang Tuukka Petaja Federico Bianchi Markku Kulmala Aerosol and Haze Laboratory Beijing Advanced Innovation Center for Soft Matter Science and EngineeringBeijing University of Chemical TechnologyBeijingChina Institute for Atmospheric and Earth System Research/Physics Faculty of ScienceUniversity of HelsinkiHelsinkiFinland State Key Joint Laboratory of Environment Simulation and Pollution Control State Environmental Protection Key Laboratory of Sources and Control of Air Pollution ComplexSchool of EnvironmentTsinghua UniversityBeijingChina State Key Joint Laboratory of Environment Simulation and Pollution Control Research Center for Eco-Environmental SciencesChinese Academy of SciencesBeijingChina Department of Environmental Science&Engineering Fudan UniversityShanghaiChina State Key Laboratory for Structural Chemistry of Unstable and Stable Species Beijing National Laboratory for Molecular SciencesInstitute of ChemistryChinese Academy of SciencesBeijingChina Jiangsu Key Laboratory of Atmospheric Environment Monitoring and Pollution Control Nanjing University of Information Science&TechnologyNanjingChina State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry Institute of Atmospheric PhysicsChinese Academy of SciencesBeijingChina Aerodyne Research Inc. BillericaMassachusettsUSA Institute of Physics University of TartuTartuEstonia Joint International Research Laboratory of Atmospheric and Earth System Sciences School of Atmospheric SciencesNanjing UniversityNanjingChina

Due to profound impact on climate and human health,air quality has attracted attention from all levels of the civil *** key step in the provision of required tools for the society to tackle the complex air quality problem is to characterize it in a comprehensive manner with a long-term ***,we describe a continuous and comprehensive observation station and its accompanying state-ofthe-art instrumentation that was established to investigate the complex urban atmospheric environment in a rapidly developing Chinese *** station,located in downtown Beijing,aims to study air quality by identifying the major atmospheric pollutants and key processes determining their formation and loss mechanisms.A few hundreds of parameters are continuously measured with the state-of-the-art instruments,including trace gas concentrations,aerosol particle size distributions,and mass concentrations,covering aerosol particle chemical composition from molecules to micrometer-sized aerosol *** produced long-term,comprehensive big data with around 1�10^(11)bytes per *** this paper,we provide an overview on the facilities of the station,the instrumentation used,the workflow of continuous observations and examples of results from 2018 to 2019 and a basis for establishing a modern long-term,comprehensive atmospheric urban observation station in other megacities.

关键词： Continuous and comprehensive atmospheric observation air pollution complex photochemical smog haze new particle formation

来源：评论

学校读者我要写书评

暂无评论

Multi2human: Controllable Human Image Generation with Multimodal Controls

SSRN

引用

SSRN 2023年

作者： Gu, Xiaoling Xu, Shengwenzhuo Wong, Yongkang Wu, Zizhao Yu, Jun Fan, Jianping Kankanhalli, Mohan S. Key Laboratory of Complex Systems Modeling and Simulation School of Computer Science and Technology Hangzhou Dianzi University China National University of Singapore Singapore School of Digital Media Technology Hangzhou Dianzi University China

Generating high-quality and diverse human images presents a substantial difficulty within the field of computer vision, especially in developing controllable generative models that can utilize input from various modalities. Such models could enable innovative applications like digital human, fashion design, and content creation. In this study, we introduce Multi2Human, a two-stage image synthesis framework for controllable human image generation with multimodal controls. In the first stage, a novel WAvelet-vqVaE (WAVE) architecture is designed to embed human images using a learnable codebook. The WAVE model enhances the conventional Vector Quantized Variational Autoencoder (VQVAE) by integrating wavelets throughout the encoder, thereby enhancing the quality of image reconstruction and synthesis. In the second stage, a new Multimodal Conditioned Diffusion Model (MCDM) is de- signed to estimate the underlying prior distribution within the discrete latent space using a discrete diffusion process, thus allowing for human image generation condi- tioned on multimodal controls. Quantitative and qualitative analysis demonstrates that the proposed method has the ability to create high-quality, lifelike full-body human images while satisfying the specified multimodal controls. Our code is available at https://***/gxl-groups/Multi2Human. © 2023, The Authors. All rights reserved.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Topological properties and localization transition in a one-dimensional non-Hermitian lattice with a slowly varying potential

引用

Physical Review B 2022年第10期105卷 104201-104201页

作者： Siqing Li Mengyao Li Yi Gao Peiqing Tong Center for Quantum Transport and Thermal Energy Science Jiangsu Key Laboratory on Opto-Electronic Technology School of Physics and Technology Nanjing Normal University Nanjing 210023 China Jiangsu Key Laboratory for Numerical Simulation of Large Scale Complex Systems School of Physics and Technology Nanjing Normal University Nanjing 210023 China

We explore the eigenenergy distribution, localization transition, and topological property of a one-dimensional non-Hermitian lattice with a slowly varying potential, which does not possess parity-time (PT) symmetry. We found that, in contrast to PT-symmetric systems whose eigenenergies can stay real until the PT-symmetry breaking point, the eigenenergies in the present system are fully complex and their distribution in the complex plane depends on the parameters. As non-Hermiticity is introduced into the system, also depending on the parameters, the eigenstates may show a crossover behavior from fully extended to fully localized as non-Hermiticity is increased, or there will be a sharp localization transition at finite non-Hermiticity. Nevertheless, this non-Hermitian system is proven to be topologically trivial, thus we conclude that in general non-Hermitian systems, the localization transition is not always of a topological nature.

关键词： Anderson localization Topological phase transition Non-Hermitean systems PT-symmetry

来源：评论

学校读者我要写书评

暂无评论

The Application of Square-Root Cubature Kalman Filter in the SINS/CNS integrated navigation system

The Application of Square-Root Cubature Kalman Filter in the...

引用

2016 IEEE Chinese Guidance, Navigation and Control Conference (IEEE CGNCC2016)

作者： Dongyang Zhang Zhihong Deng Bo Wang Mengyin Fu School of Automation and National Key Laboratory of Intelligent Control and Decision of Complex Systems Beijing Institute of Technology Nanjing University of Science and Technology

ISBN: (纸本)9781467383196

The integrated navigation system can improve the navigation accuracy utilizing the redundant *** Extended Kalman Filter(EKF) is commonly used in the integrated navigation *** usually takes only one-order of Taylor expansion and needs to calculate the Jacobian matrix,which will affect the accuracy and numerical stability of the system *** in this paper the application of Square-Root Cubature Kalman Filter(SCKF) method was proposed to solve the above problems in the SINS/CNS integrated navigation *** simulation results show that the position,velocity and attitude errors are reduced effectively compared with the EKF *** SCKF method is more suitable for the state estimation problems in integrated navigation system.

关键词： The Application of Square-Root Cubature Kalman Filter in the SINS/CNS integrated navigation system Kalman CNS Filter

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：