检索结果-南通市图书馆

Dropout training for SVMs with data augmentation

Frontiers of Computer Science 2018年第4期12卷 694-713页

作者： Ning chen Jun ZHU jianfei chen Ting chen MOE Key lab of Bioinformatics Bioinformatics Division and Center for Synthetic and Systems Biology TNLIST Tsinghua University Beijing 100084 China State Key Lab of Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University Beijing 100084 China

Dropout and other feature noising schemes have shown promise in controlling over-fitting by artificially corrupting the training data. Though extensive studies have been performed for generalized linear models, little has been done for support vector machines （SVMs）, one of the most successful approaches for supervised learning. This paper presents dropout training for both linear SVMs and the nonlinear extension with latent representation learning. For linear SVMs, to deal with the intractable expectation of the non-smooth hinge loss under corrupting distributions, we develop an iteratively re-weighted least square （IRLS） algorithm by exploring data augmentation techniques. Our algorithm iteratively minimizes the expectation of a re- weighted least square problem, where the re-weights are analytically updated. For nonlinear latent SVMs, we con- sider learning one layer of latent representations in SVMs and extend the data augmentation technique in conjunction with first-order Taylor-expansion to deal with the intractable expected hinge loss and the nonlinearity of latent representa- tions. Finally, we apply the similar data augmentation ideas to develop a new IRLS algorithm for the expected logistic loss under corrupting distributions, and we further develop a non-linear extension of logistic regression by incorporating one layer of latent representations. Our algorithms offer insights on the connection and difference between the hinge loss and logistic loss in dropout training. Empirical results on several real datasets demonstrate the effectiveness of dropout training on significantly boosting the classification accuracy of both linear and nonlinear SVMs.

关键词： dropout SVMs logistic regression data aug- mentation iteratively reweighted least square

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Characterization of topological phase of superlattices in superconducting circuits

引用

Chinese Physics B 2022年第8期31卷 688-694页

作者： jianfei chen Chaohua Wu Jingtao Fan Gang chen State Key Laboratory of Quantum Optics and Quantum Optics Devices Institute of Laser SpectroscopyShanxi UniversityTaiyuan 030006China Collaborative Innovation Center of Extreme Optics Shanxi UniversityTaiyuan 030006China Collaborative Innovation Center of Light Manipulations and Applications Shandong Normal UniversityJinan 250358China

The recent experimental observation of topological magnon insulator states in a superconducting circuit chain marks a breakthrough for topological physics with qubits, in which a dimerized qubit chain has been realized. Here, we extend such a dimer lattice to superlattice with arbitrary number of qubits in each unit cell in superconducting circuits, which exhibits rich topological properties. Specifically, by considering a quadrimeric superlattice, we show that the topological invariant(winding number) can be effectively characterized by the dynamics of the single-excitation quantum state through time-dependent quantities. Moreover, we explore the appearance and detection of the topological protected edge states in such a multiband qubit system. Finally, we also demonstrate the stable Bloch-like-oscillation of multiple interface states induced by the interference of them. Our proposal can be readily realized in experiment and may pave the way towards the investigation of topological quantum phases and topologically protected quantum information processing.

关键词： superconducting circuits topological phase transition edge state interface state

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Big Learning with Bayesian methods

引用

National Science Review 2017年第4期4卷 627-651页

作者： Jun Zhu jianfei chen Wenbo Hu Bo Zhang TNList Lab State Key Lab for Intelligent Technology and Systems CBICR Center Department of Computer Science and Technology Tsinghua University

The explosive growth in data volume and the availability of cheap computing resources have sparked increasing interest in Big learning, an emerging subfield that studies scalable machine learning algorithms,systems and applications with Big Data. Bayesian methods represent one important class of statistical methods for machine learning, with substantial recent developments on adaptive, flexible and scalable Bayesian learning. This article provides a survey of the recent advances in Big learning with Bayesian methods, termed Big Bayesian Learning, including non-parametric Bayesian methods for adaptively inferring model complexity, regularized Bayesian inference for improving the flexibility via posterior regularization, and scalable algorithms and systems based on stochastic subsampling and distributed computing for dealing with large-scale applications. We also provide various new perspectives on the large-scale Bayesian modeling and inference.

关键词： Big Bayesian Learning Bayesian non-parametrics regularized Bayesian inference scalable algorithms

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Metagenomic insights into the profile of antibiotic resistomes in sediments of aquaculture wastewater treatment system

引用

Journal of Environmental Sciences 2022年第3期34卷 345-355页

作者： jianfei chen Yuyin Yang Xinshu Jiang Yanchu Ke Tao He Shuguang Xie State Key Joint Laboratory of Environmental Simulation and Pollution Control College of Environmental Sciences and EngineeringPeking UniversityBeijing 100871China State Key Joint Laboratory of Environment Simulation and Pollution Control(SKJLESPC) Beijing Key Laboratory for Emerging Organic Contaminants Control(BKLEOC)School of EnvironmentPOPs Research CenterTsinghua UniversityBeijing 100084China South China Institute of Environmental Sciences(SCIES) Ministry of Ecology and Environment(MEE)Guangzhou 510655China

To meet the rapidly growing global demand for aquaculture products,large amounts of antibiotics were used in aquaculture,which might accelerate the evolution of antibiotic-resistant bacteria(ARB)and the propagation of antibiotic genes(ARGs).In our research,we revealed the ARGs profiles,their co-occurrence with mobile genetic elements(MGEs),and potential hosts in sediments of a crab pond wastewater purification system based on metagenomic *** residual antibiotic seems to increase the propagation of ARGs in the crab pond,but there was no clear relationship between a given antibiotic type and the corresponding resistance *** effect of aquaculture on sediment was not as profound as that of other anthropogentic activities,but increased the relative abundance of sulfon-amide resistance gene.A higher abundance of MGEs,especially plasmid,increased the po-tential ARGs dissemination risk in crab and purification *** and sulfonamide resistance genes had greater potential to transfer because they were more frequently carried by *** horizontal gene transfer was likely to occur among a variety of microorgan-isms,and various ARGs hosts including Pseudomonas,Acinetobacter,Escherichia,and Klebsiella were *** community influenced the composition of ARG hosts,and Pro-teobacteria was the predominant ***,our study provides novel insights into the environmental risk of ARGs in sediments of aquaculture wastewater treatment system.

关键词： Metagenomic sequencing Aquaculture Antibiotic resistome Antibiotic resistance genes(ARGs) Mobile genetic elements(MGEs) ARG hosts

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Identification of Gesture Based on Combination of Raw sEMG and sEMG Envelope Using Supervised Learning and Univariate Feature Selection

引用

Journal of Bionic Engineering 2019年第4期16卷 647-662页

作者： Shili Liang Yansheng Wu jianfei chen Ling Zhang Peipei chen Zongqian Chai Chunlei Cao School of Physics Northeast Normal UniversityChangchun 130022China College of Science Changchun University of Science and TechnologyChangchun 130022China

In this paper,we propose a novel study for gesture identification using surface electromyography(sEMG)signal,and the raw sEMG signal and the sEMG envelope signal are collected by the sensor at the same *** efficient method of gesture identification based on the combination of two signals using supervised learning and univariate feature selection is *** previous research techniques,researchers tend to use the raw sEMG signal and extract several constant features for classification,which inevitably causes a result of ignoring individual *** experiment shows that both the optimal feature set and redundant feature set are not same for different *** order to address this problem,we extract all the common features from two signals,up to 76 features,most of which has been established as the common EMG-based gesture *** addition,extracting too many features in an application can reduce operational efficiency,so we apply for feature selection to get the optimal feature set and decrease the number of extracting *** a result,the combination of two signals is better than using a single *** feature selection can be used to select optimal feature set from all features to achieve the best classification performance for each *** experimental results demonstrate that the proposed method achieves the performance with the highest accuracy of 95%for identifying up to nine gestures only using two ***,we develop a real-time intelligent sEMG-driven bionic hand system by using the proposed method.

关键词： gesture identification surface electromyography(sEMG) supervised learning univariate feature selection individual differences

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A Double Input-parallel-output-series Hybrid Switched-capacitor Boost Converter

引用

Chinese Journal of Electrical Engineering 2020年第4期6卷 15-27页

作者： jianfei chen Kewei Ding Yulin Zhong Fujin Deng Sayed Abulanwar Department of Electrical and Computer Engineering Wayne State UniversityMichigan 48202USA Broad-Ocean Technologies NoviMichigan 30120USA School of Electrical Engineering Southeast UniversityNanjing 210096China Electrical Engineering Department Mansoura UniversityMansoura 35516Egypt

A double input-parallel-output-series hybrid switched-capacitor boost(DIPOS-HSCB)converter is proposed which consists of two different kinds of input-parallel-output-series(IPOS)circuits,i.e.,inner IPOS circuit and outer IPOS *** boost modules and one switched-capacitor network build an inner IPOS circuit based IPOS-HSCB converter and two IPOS-HSCB converters develop the outer IPOS circuit based DIPOS-HSCB *** the proposed DIPOS-HSCB converter,a high voltage-gain with low component stress and small input current ripple are ***,an automatic current balancing function for all input inductor currents can be also achieved using a special carrier phase-shifted modulation scheme.A prototype rated at 200 V/120 W has been developed and the maximum efficiency of the proposed DIPOS-HSCB converter is 95% at 120 *** steady and dynamic results are presented to validate the effectiveness of the proposed DIPOS-HSCB converter.

关键词： IPOS switched-capacitor boost converter phase-shifted modulation automatic current balancing

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Wavelet method optimised by ant colony algorithm used for extracting stable and unstable signals in intelligent substations

引用

CAAI Transactions on Intelligence Technology 2022年第2期7卷 292-300页

作者： Tianyan Jiang Xiao Yang Yuan Yang Xi chen Maoqiang Bi jianfei chen School of Electrical and Electronic Engineering Chongqing University of TechnologyChongqingChina DepartmentState Key Laboratory of Power Transmission Equipment&System and New Technology Chongqing UniversityChongqingChina Department of Electrical and Computer Engineering University of MarylandCollege ParkMarylandUSA

Partial discharge(PD)signals are an important index to evaluate the operation state of intelligent *** correct distinction of PD pulse and interference pulse has become a challenging *** of the noise and the low signal-to-noise ratio,the stable signals become *** selection of a wavelet basis,the selection rule of thresholdλand the design of the threshold function are the key factors affecting the final denoising ***,an enhanced ant colony optimisition wavelet(ACOW)algorithm was applied to find the global optimal threshold through the continuous derivative threshold function and the ant colony optimisation(ACO)*** the same time the efficiency of adaptive search calculation,was also significantly *** method of the ACOW algorithm was compared with the soft wavelet method,gradient-based wavelet method and the genetic optimisation wavelet(GOW)*** these four methods to denoise four typical signals,different mean square errors(MSE),magnitude errors(ME)and time costs were ***,the results show that the ACOW method can achieve the minimum MSE and has less time *** generates significantly smaller waveform distortion than the other three threshold estimation *** addition,the high efficiency and good quality of the output signals are beneficial to the diagnosis of local discharge signals in intelligent substations.

关键词： algorithm noise method

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

What have we known so far about microplastics in drinking water treatment?A timely review

引用

Frontiers of Environmental Science & Engineering 2022年第5期16卷 39-56页

作者： Jinkai Xue Seyed Hesam-Aldin Samaei jianfei chen Ariana Doucet Kelvin Tsun Wai Ng Environmental Systems Engineering Faculty of Engineering&Applied ScienceUniversity of Regina3737 Wascana ParkwayRegina SK S4S 0A2Canada

Microplastics(MPs)have been widely detected in drinking water sources and tap water,raising the concern of the effectiveness of drinking water treatment plants(DWTPs)in protecting the public from exposure to MPs through drinking *** collected and analyzed the available research articles up to August 2021 on MPs in drinking water treatment(DWT),including laboratory-and full-scale *** article summarizes the major MP compositions(materials,sizes,shapes,and concentrations)in drinking water sources,and critically reviews the removal efficiency and impacts of MPs in various drinking water treatment *** discussed drinking water treatment processes include coagulation-flocculation(CF),membrane filtration,sand filtration,and granular activated carbon(GAC)*** DWT processes that are purposed for particle removal are generally effective in reducing MPs in *** influential factors to MP removal are discussed,such as coagulant type and dose,MP material,shape and size,and water *** is anticipated that better MP removal can be achieved by optimizing the treatment ***,the article framed the major challenges and future research directions on MPs and nanoplastics(NPs)in DWT.

关键词： Microplastics Drinking water treatment Coagulation Flocculation Membrane Filtration

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Nanostructured BiVO4 Derived from Bi-MOF for Enhanced Visible-light Photodegradation

引用

Chemical Research in Chinese Universities 2020年第1期36卷 120-126页

作者： chen jianfei chen Xiaoyu ZHANG Xing YUAN Yao BI Ruyi YOU Feifei WANG Zumin YU Ranbo Department of Physical Chemistry School of Metallurgical and Ecological EngineeringUniversity of Science and Technology BeijingBeijing 100083P.R.China State Key Laboratory of Biochemical Engineering Institute of Process EngineeringChinese Academy of SciencesBeijing 100190P.R.China Key Laboratory of Advanced Material Processing&Mold Ministry of EducationZhengzhou UniversityZhengzhou 450002P.R.China

BiVO4,a promising visible-light responding photocatalyst,has aroused extensive research interest because of inexpensiveness and excellent chemical ***,its main drawback is the poor photoinduced charge-transfer *** nanostructures is an effective way to tackle this ***,we put forward a new method to prepare nanostructured BiVO4 from Bi-based metal-organic frameworks[Bi-MOF(CAU-17)]*** as-prepared material has a rod-like morphology inherited from the Bi-MOF sacrificial template and consists of small nanoparticle as building *** with its counterparts prepared by conventional methods,MOF-derived nanostructured BiVO4 shows better light absorption ability,narrower bandgap,and improved electrical conductivity as well as reduced ***,BiVO4 nanostructure demonstrates high photocatalytic activity under visible light towards the degradation of methylene *** blue can be degraded up to 90%within 30 min with a reaction rate constant of 0.058 ***,the cycling stability of the catalyst is excellent to withstand unchanged degradation efficiency for at least 5 cycles.

关键词： BiVO4 Nanostructure Metal-organic framework Ternary metal oxide Photocatalysis

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A comparative research on built-land changes between Fujian and Taiwan provinces

引用

Journal of Geographical Sciences 2007年第2期17卷 245-256页

作者： WEI Suqiong chen jianfei College of Geographical Sciences Fujian Normal University Fuzhou 350007 China College of Geographical Sciences Guangzhou University Guangzhou 510006 China

Comparative study of LUCC between Fujian and Taiwan provinces could be a good case for study because of their geographical proximities and historical, cultural similarities. Fujian and Taiwan had gone through and currently are in different economic development stages. Taiwan is in the stage of post-industrialization and Fujian is in the intermediate stage of industrialization. By using the official statistics, changes of built-land in Fujian and Taiwan were analyzed in details. The results showed that these two provinces seemed to have the same characteristics of built-land changes, i.e., accelerating development, and concentrated distribution on both sides of the Taiwan Straits, forming a pattern with the economic layout of two sides acting in cooperation with eath other. If comparing different industrialized areas in Fujian with industrialized stages in Taiwan, it is clear that these two provinces seemed to have the same characteristics of synthesized index of built-land change （Lc）, i.e., which is from stabilization or comparative stabilization at the initial stages of industrialization to high expansion at the intermediate stages of industrialization, then followed by low consumption at later stages of industrialization, but the Lc of the initial and the later stages of industrialization in Taiwan were slightly higher than those in Fujian. The results from comparison on relevant indexes have shown： The industrial structure changing rate of Fujian is actually higher than that of built-land in Taiwan, but its using efficiency of regional built-land is also obviously lower than that in Taiwan in the intermediate and later stages of industrialization of Fujian. The major driving forces of the built-land changes in both provinces were analyzed in Causal Models of Path Analysis, and its result indicated that the difference in built-land changes of the two provinces is closely related to their economic development stages and industrial structure. Countermeasure for the

关键词： built-land change regional comparison industrialization Fujian Taiwan

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：