检索结果-内蒙古大学图书馆

作者： Kun Zhou Wen Yong Wang Teng Hu Chen Huang Wu School of Computer Science and Engineering University of Electronic Science and Technology of China Institute of Computer Application China Academy of Engineering Physics School of Mathematics and Finance Putian University

Time series Forecasting(TSF) has been a research hotspot and widely applied in many areas such as financial, bioinformatics, social sciences and engineering. This article aimed at comparing the forecasting performances using the traditional Auto-Regressive Integrated Moving Average(ARIMA) model with the deep neural network model of Long Short Term Memory(LSTM) with attention mechanism which achieved great success in sequence modelling. We first briefly introduced the basics of ARIMA and LSTM with attention models, summarized the general steps of constructing the ARIMA model for the TSF task. We obtained the dataset from Kaggle competition web traffic and modelled them as TSF problem. Then the LSTM with attention mechanism model was proposed to the TSF. Finally forecasting performance comparisons were conducted using the same dataset under different evaluation metrics. Both models achieved comparable results with the up-to-date methods and LSTM slightly outperformed the classical counterpart in TSF task.

关键词： Comparison of Time Series Forecasting Based on Statistical ARIMA Model and LSTM with Attention Mechanism Model

来源：评论

学校读者我要写书评

暂无评论

CAT: Cross-attention transformer for one-shot object detection

arXiv

引用

arXiv 2021年

作者： Lin, Weidong Deng, Yuyan Gao, Yang Wang, Ning Zhou, Jinghao Liu, Lingqiao Zhang, Lei Wang, Peng School of Computer Science Northwestern Polytechnical University China National Engineering Lab for Integrated Aero-Space-Ground-Ocean Big Data Application Technology China University of Adelaide Australia

Given a query patch from a novel class, one-shot object detection aims to detect all instances of that class in a target image through the semantic similarity comparison. However, due to the extremely limited guidance in the novel class as well as the unseen appearance difference between query and target instances, it is difficult to appropriately exploit their semantic similarity and generalize well. To mitigate this problem, we present a universal Cross-Attention Transformer (CAT) module for accurate and efficient semantic similarity comparison in one-shot object detection. The proposed CAT utilizes transformer mechanism to comprehensively capture bi-directional correspondence between any paired pixels from the query and the target image, which empowers us to sufficiently exploit their semantic characteristics for accurate similarity comparison. In addition, the proposed CAT enables feature dimensionality compression for inference speedup without performance loss. Extensive experiments on COCO, VOC, and FSOD under one-shot settings demonstrate the effectiveness and efficiency of our method, e.g., it surpasses CoAE, a major baseline in this task by 1.0% in AP on COCO and runs nearly 2.5 times faster. Code will be available in the future. Copyright © 2021, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Correction: Shielding effectiveness performance of polyaniline-NiFe2O4:Cu composites for sub-8 GHz applications

引用

Optical and Quantum Electronics 2023年第9期55卷 1-2页

作者： Şahin, Ethem İlhan Emek, Mehriban Ibrahim, Jamal Eldin F. M. Yumuşak, Görkem Kartal, Mesut Advanced Technology Research and Application Center Adana Alparslan Türkeş Science and Technology University Adana Turkey Department of Physics Computer Technologies Golbasi Vocational School Adiyaman University Gölbaşi Adıyaman Turkey Department of Materials Technology University of Bahri Khartoum Sudan Institute of Ceramic and Polymer Engineering University of Miskolc Egyetemvaros Miskolc Hungary Department of Metallurgical and Materials Engineering Marmara University Göztepe Istanbul Turkey Department of Electronics and Communication Engineering İstanbul Technical University Maslak Istanbul Turkey

来源：评论

学校读者我要写书评

暂无评论

Randomized Latent Factor Model for High-dimensional and Sparse Matrices from Industrial applications

引用

IEEE/CAA Journal of Automatica Sinica 2019年第1期6卷 131-141页

作者： Mingsheng Shang Xin Luo Zhigang Liu Jia Chen Ye Yuan MengChu Zhou the Chongqing Engineering Research Center of Big Data Application for Smart Cities and Chongqing Key Laboratory of Big Data and Intelligent ComputingChongqing Institute of Green and Intelligent TechnologyChinese Academy of Sciences the School of Computer Science and Engineering Beihang University the Department of Electrical and Computer Engineering New Jersey Institute of Technology

Latent factor(LF) models are highly effective in extracting useful knowledge from High-Dimensional and Sparse(HiDS) matrices which are commonly seen in various industrial applications. An LF model usually adopts iterative optimizers,which may consume many iterations to achieve a local optima,resulting in considerable time cost. Hence, determining how to accelerate the training process for LF models has become a significant issue. To address this, this work proposes a randomized latent factor(RLF) model. It incorporates the principle of randomized learning techniques from neural networks into the LF analysis of HiDS matrices, thereby greatly alleviating computational burden. It also extends a standard learning process for randomized neural networks in context of LF analysis to make the resulting model represent an HiDS matrix *** results on three HiDS matrices from industrial applications demonstrate that compared with state-of-the-art LF models, RLF is able to achieve significantly higher computational efficiency and comparable prediction accuracy for missing data.I provides an important alternative approach to LF analysis of HiDS matrices, which is especially desired for industrial applications demanding highly efficient models.

关键词： Big data high-dimensional and sparse matrix latent factor analysis latent factor model randomized learning

来源：评论

学校读者我要写书评

暂无评论

Improve the Performance of LSM-Tree based on Key-Value via Multithreading 20

Improve the Performance of LSM-Tree based on Key-Value via M...

引用

20th International Symposium on Distributed Computing and applications for Business engineering and science, DCABES 2021

作者： Gao, Yuan Xie, Ping Hua, Wendi Lv, Meng Lu, Jiating College of Computer of Qinghai Normal University Xining810016 China State Key Laboratory of Tibetan Intelligent Information Processing and Application Xining810016 China Key Laboratory of Internet of Things of Qinghai Province Xining810016 China Academy of Plateau Science and Sustainability Xining810016 China School of Physical and Electrical Engineering Qinghai Normal University Xining810016 China

ISBN: (纸本)9781665428897

In the era of big data, key-value storage systems based on Log-Structure Merge tree (LSM-tree) are widely used in numerous industries. LSM-Tree is divided into two parts, one part is in the memory and the other part is on the hard disk. When writing data, first write to the memory, when the Immutable Memtable in the memory is full of data, then write to the disk to output a Sorted String Table (SSTable) file. However, the use of writing data from memory to disk is single-threaded. When the Immutable Memtable writes data to the disk, it will block other threads. This article uses multiple Immutable Memtables and multi-threaded writing to solve the single-thread blocking problem. By running the benchmarks of LevelDB to analyze the experiment, the experimental results indicate a higher improvement. © 2021 IEEE.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

FEDLAB: A FLEXIBLE FEDERATED LEARNING FRAMEWORK

arXiv

引用

arXiv 2021年

作者： Zeng, Dun Liang, Siqi Hu, Xiangjing Wang, Hui Xu, Zenglin School of Computer Science and Engineering University of Electronic Science and Technology of China China Rich Media Big Data Analytics and Application Key Laboratory Shenzhen Research Institute The Chinese University of Hong Kong Hong Kong School of Computer Science and Technology Harbin Institute of Technology Shenzhen China Peng Cheng Laboratory Shenzhen China

Federated learning (FL) is a machine learning field in which researchers try to facilitate model learning process among multiparty without violating privacy protection regulations. Considerable effort has been invested in FL optimization and communication related researches. In this work, we introduce FedLab, a lightweight open-source framework for FL simulation. The design of FedLab focuses on FL algorithm effectiveness and communication efficiency. Also, FedLab is scalable in different deployment scenario. We hope FedLab could provide flexible API as well as reliable baseline implementations, and relieve the burden of implementing novel approaches for researchers in FL community. The source code is available at https://***/SMILELab-FL/FedLab. © 2021, CC BY.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Impact of Structural Properties on Network Structure for Online Social Networks

引用

Procedia computer science 2020年 167卷 1200-1209页

作者： Subhayan Bhattacharya Sankhamita Sinha Sarbani Roy School of Mobile Computing and Communication Jadavpur University Kolkata Department of Masters of Computer Applications Meghnad Saha Institute of Technology Kolkata Department of Computer Science and Engineering Jadavpur University Kolkata

Online Social Networks (OSNs) are largely popular. People interact daily on networks such as Facebook, Twitter, YouTube, Email, Messenger, WhatsApp, Google+, Quora, LiveJournal etc. and the types of interaction on these networks are different. The types of content that people share on these networks are also different. Each of the OSNs serves a different purpose, such as sharing multimedia, microblogging, serving as a Question and Answer forum etc. Moreover, the community structure of these networks is also different. However, currently, all OSNs are considered as scale-free network based on the power-law degree distribution nature of these networks. In this paper, the effect of network properties such as density, diameter, degree distribution, global clustering coefficient, local clustering coefficient, homophily, assortativity and other centrality measures such as power-law exponent, eigenvector centrality and closeness centrality on seven online social networks, namely, Facebook, Twitter, YouTube, Email, Google+, Epinions and Gowalla networks are studied. The differences and similarities in the empirical results are analyzed, and the OSNs are divided categories based on the observed differences and similarities in network properties and centrality measures.

关键词： Online Social Networks Categorisation Centrality Measures Assortativity Homophily Clustering Coefficient

来源：评论

学校读者我要写书评

暂无评论

Sputtered Sn-doped Ga2O3 films under balance controlled of energy supply and ion bombardment for solar-blind detection application

引用

Vacuum 2024年 225卷

作者： Wang, Chen Fan, Wei-Hang Cao, Rong-Jun Fan, Hui-Chen Xu, Yi-Hong Kang, Pin-Chun Wu, Ting-Zhu Kuo, Hao-Chung Lien, Shui-Yang Zhu, Wen-Zhang Xiamen Key Laboratory of Development and Application for Advanced Semiconductor Coating Technology School of Opto-electronic and Communication Engineering Xiamen University of Technology Xiamen361024 China National Measurement and Testing Center for Flat Panel Display Industry Xiamen Institute of Measurement and Testing Xiamen361004 China National Integrated Circuit Industry and Education Integration Innovation Platform Department of Electronic Science Xiamen University Xiamen361005 China Department of Photonics & Graduate Institute of Electro-Optical Engineering College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan Department of Biomedical Engineering Da-Yeh University Changhua51591 Taiwan

The characteristics of amorphous Sn-doped Ga2O3 films deposited using radio frequency magnetron sputtering at room temperature under different sputter powers have been demonstrated. A balance mechanism of energy supply and ion bombardment controlled by sputter power has been found during the deposition. The increasing growth rate as power can be due to the increased yield sputtering species indicated by an in situ optical emission spectroscopy. As the power increases to 700 W, an obvious change from discrete nanoparticles to flat-continuous film can be observed because of sufficient energy supply for oxidation reaction, which leads to the maximum ratio of lattice oxygen and Sn4+ indicating the optimal film quality and conductive property. However, a non-negligible high energy ion bombardment presented at 900 W leads to the degraded film quality. The influence of sputtering power on Sn-doped Ga2O3 solar-blind photodetectors is demonstrated. It can obtain a good performance with an ultra-low dark current density of 2.2 × 10−11 A/cm2, a substantial light-dark current ratio of 1.08 × 103 and fast rise/decay time of 1.59 s/0.54 s at 700 W. The revelation of the role of sputtering power on the properties of Sn-doped Ga2O3 films may offer an interesting direction for low-cost and high-performance solar-blind photodetectors. © 2024 Elsevier Ltd

关键词： Gallium compounds

来源：评论

学校读者我要写书评

暂无评论

A Dynamics Theory of Implicit Regularization in Deep Low-Rank Matrix Factorization

arXiv

引用

arXiv 2022年

作者： Cao, Jian Qian, Chen Huang, Yihui Chen, Dicheng Gao, Yuncheng Dong, Jiyang Guo, Di Qu, Xiaobo Department of Electronic Science Biomedical Intelligent Cloud R&D Center Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance Xiamen University Xiamen361005 China School of Computer and Information Engineering Fujian Engineering Research Center for Medical Data Mining and Application Xiamen University of Technology Xiamen361024 China

Implicit regularization is an important way to interpret neural networks. Recent theory starts to explain implicit regularization with the model of deep matrix factorization (DMF) and analyze the trajectory of discrete gradient dynamics in the optimization process. These discrete gradient dynamics are relatively small but not infinitesimal, thus fitting well with the practical implementation of neural networks. Currently, discrete gradient dynamics analysis has been successfully applied to shallow networks but encounters the difficulty of complex computation for deep networks. In this work, we introduce another discrete gradient dynamics approach to explain implicit regularization, i.e. landscape analysis. It mainly focuses on gradient regions, such as saddle points and local minima. We theoretically establish the connection between saddle point escaping (SPE) stages and the matrix rank in DMF. We prove that, for a rank-R matrix reconstruction, DMF will converge to a second-order critical point after R stages of SPE. This conclusion is further experimentally verified on a low-rank matrix reconstruction problem. This work provides a new theory to analyze implicit regularization in deep learning. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Improved hybrid query tree anti-collision algorithm 8th

Improved hybrid query tree anti-collision algorithm

引用

8th International Conference on computer engineering and Networks, CENet 2018

作者： Deng, Hongwei Li, Lang Liang, Xiaoman School of Computer Science and Technology Hengyang Normal University HengyangHunan421002 China School of Information Science and Engineering Central South University ChangshaHunan410083 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002 China

ISBN: (纸本)9783030146795

An improved algorithm based on the hybrid query tree (HQT) algorithm is proposed in this work. Tags are categorized according to the combined information of the highest bit of collision and second-highest bit of collision. Then, these tags are used to decide the postponement of several delayed-response time slots, to diminish the collision probability. The performance analysis results show that this algorithm is superior to the query tree (QT) and HQT algorithms. It reduces the query time and system communication traffic and improves the tag identification efficiency. © Springer Nature Switzerland AG 2020.

关键词： Trees (mathematics)

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：