检索结果-南通市图书馆

GPS: graph contrastive learning via multi-scale augmented views from adversarial pooling

Science China(Information Sciences) 2025年第1期68卷 145-158页

作者： Wei JU Yiyang GU Zhengyang MAO Ziyue qiao Yifang QIN Xiao LUO Hui XIONG Ming ZHANG School of Computer Science National Key Laboratory for Multimedia Information ProcessingPeking University Artificial Intelligence Thrust The Hong Kong University of Science and Technology Department of Computer Science University of California

Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by maximizing agreement between original graphs and their augmented views(i.e., positive views). Unfortunately, these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts. Moreover, these strategies may fail to generate challenging positive views to provide sufficient supervision signals. In this paper, we present a novel approach named graph pooling contrast(GPS) to address these *** by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy, we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics, i.e., strongly-augmented view and weakly-augmented view. Then, we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning, where our pooling module is adversarially trained with respect to the encoder for adversarial robustness. Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

关键词： graph representation learning graph neural networks graph contrastive learning graph augmentations graph pooling

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

Weierstrass Type Representation of Willmore Surfaces in S^n

引用

Acta Mathematica Sinica,English Series 2004年第6期20卷 1029-1046页

作者： qiao ling XIA Yi Bing SHEN Department of Mathematics Zhejiang University

In this paper,we reformulate the Euler-Lagrange equations of Willmore surfaces in S^n as the flatness of a family of certain loop algebra-valued *** we can give the Weierstrass type representation of conformal Willmore *** also discuss the relations between conformal Willmore surfaces in S^n and minimal surfaces in constant curvature spaces S^n,R^n,H^n,and prove that some special Willmore surfaces can be derived from minimal surfaces in S^n,R^n,H^n.

关键词： Willmore surfaces Extended lift of a conformal Willmore immersion Loop group Weierstrass type representation Minimal surface

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

Morita Contexts and Galois Theory for Weak Hopf Comodulelike Algebras

引用

Acta Mathematica Sinica,English Series 2011年第4期27卷 757-772页

作者： qiao ling GUO Shuan Hong WANG College of Mathematics and Information Engineering Jiaxing University Jiaxing 314001 P. R. China Department of Mathematics Southeast University Nanjing 210096 P. R. China

We investigate the Morita context and graded cases for weak group corings and derive some equivalent conditions for p to be surjective. Furthermore, we develop Galois theory for weak group corings. As an application, ... 详细信息

关键词： Weak group corings graded Mortira contexts weak Hopf comodulelike algebras

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

Effective ensemble learning approach for SST field prediction using attention-based PredRNN

引用

Frontiers of Computer Science 2023年第1期17卷 143-155页

作者： Baiyou qiao Zhongqiang WU ling MA Yicheng Zhou Yunjiao SUN School of Computer Science and Engineering Northeastern UniversityShenyang 110169China Key Laboratory of Intelligent Computing in Medical Image Ministry of EducationNortheastern UniversityShenyang 110169China

Accurate prediction of sea surface temperature (SST) is extremely important for forecasting oceanic environmental events and for ocean studies. However, the existing SST prediction methods do not consider the seasonal periodicity and abnormal fluctuation characteristics of SST or the importance of historical SST data from different times;thus, these methods suffer from low prediction accuracy. To solve this problem, we comprehensively consider the effects of seasonal periodicity and abnormal fluctuation characteristics of SST data, as well as the influence of historical data in different periods, on prediction accuracy. We propose a novel ensemble learning approach that combines the Predictive Recurrent Neural Network(PredRNN) network and an attention mechanism for effective SST field prediction. In this approach, the XGBoost model is used to learn the long-period fluctuation law of SST and to extract seasonal periodic features from SST data. The exponential smoothing method is used to mitigate the impact of severely abnormal SST fluctuations and extract the a priori features of SST data. The outputs of the two aforementioned models and the original SST data are stacked and used as inputs for the next model, the PredRNN network. PredRNN is the most recently developed spatiotemporal deep learning network, which simulates both spatial and temporal representations and is capable of transferring memory across layers and time steps. Therefore, we used it to extract the spatiotemporal correlations of SST data and predict future SSTs. Finally, an attention mechanism is added to capture the importance of different historical SST data, weigh the output of each step of the PredRNN network, and improve the prediction accuracy. The experimental results on two ocean datasets confirm that the proposed approach achieves higher training efficiency and prediction accuracy than the existing SST field prediction approaches do.

关键词： SST prediction ensemble learning XGBoost PredRNN attention mechanism

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

An Enhanced Dual-Layer Ensemble Framework for Porosity Prediction Utilizing Conventional Well Logging Data

引用

Journal of Earth Science 2025年

作者： Lu qiao Shengyu Yang Qinhong Hu Huijun Wang Taohua He National Key Laboratory of Deep Oil and Gas， China University of Petroleum (East China) Laboratory for Marine Mineral Resources， Qingdao National Laboratory for Marine Science and Technology Hubei Key Laboratory of Petroleum Geochemistry and Environment (Yangtze University)， Yangtze University

Porosity is the one of the most fundamental and critical parameters for reservoir characterization and is closely related to the hydrocarbon storage capacity. The interpretation methods of conventional well logs have poor generalization capabilities to achieve the desired accuracy in porosity prediction. The increasing adoption of machine learning has ameliorated this predicament to some extent， but it remains difficult to reconstruct the complex and nonlinear relationship between logging responses and porosity by relying on individual machine learning models alone. This work proposes a two-layer model ensemble framework to predicte porosity， which includes base regressors in the first layer and meta regressor in the second layer. The results indicate that the Light Gradient Boosting Machine (LightGBM) - Categorical-Features Gradient Boosting (CatBoost) ensemble model provides the highest prediction accuracy for the given dataset and is applied for porosity evaluation at other logged depths. This work also points out that the base regressor should be accurate and complex， while the structure of the meta-regressor should be simple in order to achieve the desired performance. Furthermore， the proposed model ensemble framework is versatile and applicable to reservoir evaluation， engineering development， and other potential fields due to its effectiveness and convenience.

关键词：

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

基于k-center聚类和最近邻中心的公平数据汇总

引用

西华师范大学学报（自然科学版） 2025年第1期46卷 95-103页

作者：何艳黄巧玲郑伯川西华师范大学数学与信息学院四川南充637009 西华师范大学计算机学院四川南充637009

公平数据汇总是指从每种数据类别中选择有代表性的子集,且满足公平性要求。在大数据时代,每种类别的数据都是海量的,因此公平数据汇总研究具有非常重要的现实意义。为了使公平数据汇总的数据点更具有代表性,提出了基于k-center聚类和最... 详细信息

公平数据汇总是指从每种数据类别中选择有代表性的子集,且满足公平性要求。在大数据时代,每种类别的数据都是海量的,因此公平数据汇总研究具有非常重要的现实意义。为了使公平数据汇总的数据点更具有代表性,提出了基于k-center聚类和最近邻中心的公平数据汇总算法。算法主要包括2个基本步骤:(1)通过k-center聚类,将k个簇中心作为当前汇总结果;(2)选择满足公平约束的原簇中心的最近邻点作为新簇中心。由于更新簇中心时选择的是原簇中心的最近邻点,因此相对随机选择的数据点,最近邻点更具有代表性,是除原始簇中心外的次优代表点。同时,寻找最近邻点作为新的簇中心能最大限度减少公平代价。在2个模拟数据集和6个UCI真实数据集上的对比实验结果表明,所提出的算法在近似因子和公平代价方面都优于对比算法,说明所提出的算法获得的数据汇总更具有代表性。

关键词：最近邻点 k-center聚类数据汇总公平约束

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

Plasmonic tandem heterojunctions enable high-efficiency charge transfer for broad spectrum photocatalytic hydrogen production

引用

Journal of Energy Chemistry 2025年第1期100卷 710-720页

作者： Wenxuan Chen Xiu-Qing qiao Guo Hui Bojing Sun Dongfang Hou Meidi Wang Xueqian Wu Tao Wu Dong-Sheng Li College of Materials and Chemical Engineering Key Laboratory of Inorganic Nonmetallic Crystalline and Energy Conversion Materials China Three Gorges University Hubei Three Gorges Laboratory College of Chemistry and Materials Science and Guangdong Provincial Key Laboratory of Supramolecular Coordination Chemistry Jinan University

Rational engineering of semiconductor photocatalysts for efficient hydrogen production is of great significance but still challenging,primarily due to the limitations in charge transfer ***,a fascinating plasmonic tandem heterojunction with the hc-CdS/Mo2C@C heterostructure is aimfully prepared for effectively promoting the charge separation kinetics of the CdS photocatalyst via the synergistic strategy of phase junction,Schottky junction,and photothermal *** difference in atomic configuration between cubic-CdS （c-CdS） and hexagonal-CdS （h-CdS） leads to effective charge separation through a typical Ⅱ charge transfer mechanism,and plasmonic Schottky junction further extracts the electrons in the hc-CdS phase junction to realize gradient charge ***,the photothermal effect of Mo2C@C helps to expand the light absorption,accelerate charge transfer kinetics,and reduce the hydrogen evolution energy *** carbon layer provides a fast channel for charge transfer and protects the photocatalyst from *** a result,the optimized hc-CMC photocatalyst exhibits a significantly high photocatalytic H2production activity of 28.63 mmol/g/h and apparent quantum efficiency of 61.8%,surpassing most of the reported *** study provides a feasible strategy to enhance the charge transfer kinetics and photocatalytic activity of CdS by constructing plasmonic tandem heterogeneous junctions.

关键词： Cadmium sulfide

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

Small extracellular vesicles derived from human induced pluripotent stem cell-differentiated neural progenitor cells mitigate retinal ganglion cell degeneration in a mouse model of optic nerve injury

引用

Neural Regeneration Research 2025年第2期20卷 587-597页

作者： Tong Li Hui-Min Xing Hai-Dong Qian qiao Gao Sheng-Lan Xu Hua Ma Zai-Long Chi State Key Laboratory of Ophthalmology Optometry and Visual ScienceEye HospitalWenzhou Medical UniversityWenzhouZhejiang ProvinceChina National Clinical Research Center for Ocular Diseases Eye HospitalWenzhou Medical UniversityWenzhouZhejiang ProvinceChina

Several studies have found that transplantation of neural progenitor cells(NPCs)promotes the survival of injured ***,a poor integration rate and high risk of tumorigenicity after cell transplantation limits their clinical *** extracellular vesicles(sEVs)contain bioactive molecules for neuronal protection and *** studies have shown that stem/progenitor cell-derived sEVs can promote neuronal survival and recovery of neurological function in neurodegenerative eye diseases and other eye *** this study,we intravitreally transplanted sEVs derived from human induced pluripotent stem cells(hiPSCs)and hiPSCs-differentiated NPCs(hiPSC-NPC)in a mouse model of optic nerve *** results show that these intravitreally injected sEVs were ingested by retinal cells,especially those localized in the ganglion cell *** with hiPSC-NPC-derived sEVs mitigated optic nerve crush-induced retinal ganglion cell degeneration,and regulated the retinal microenvironment by inhibiting excessive activation of *** analysis further revealed that hiPSC-NPC derived sEVs transported neuroprotective and anti-inflammatory miRNA cargos to target cells,which had protective effects on RGCs after optic nerve *** findings suggest that sEVs derived from hiPSC-NPC are a promising cell-free therapeutic strategy for optic neuropathy.

关键词： exosome miRNA neural progenitor cell neurodegeneration neuroinflammation neuroprotection optic nerve crush optic neuropathy retinal ganglion cell small extracellular vesicles

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

Oxidation resistance 1 （OXR1） participates in silkworm defense against bacterial infection through the JNK pathway

引用

Insect Science 2017年第1期24卷 17-26页

作者： Li-De Su qiao-ling Zhang Zhiqiang Lu Department of Entomology College of Plant Protection Ministry of Education Northwest A &F University Yangling Shaanxi 712100 China Key Laboratory of Plant Protection Resources and Pest Management Ministry of Education Northwest A &F University Yangling Shaanxi 712100 China

Bacterial infection causes enhanced reactive oxygen species （ROS） levels in insects. Oxidation resistance 1 （OXR1） plays an antioxidant role in eukaryotic organisms, including insects. In this report, we demonstrated that Pseudomonas aeruginosa and Staphylococcus aureus infection and hydrogen peroxide （H2O2） injection induced the expression of specific transcriptional isoforms of OXR1 in larval silkworms. We further showed that a Jun kinase （JNK） pathway inhibitor, SP600125, down-regulated expression of OXRI during infection, leading to elevated H2O2 levels in the hemolymph, resulting in lower viability of the injected bacteria inside the silkworm larvae. Our study suggests that OXR1 participates in protecting larval silkworms from oxidative stress and bacterial infection through the JNK pathway.

关键词： Bombyx mori infection Jun kinase pathway oxidation resistance 1 reactive oxygen species

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

Change Rules of Γ_2 Particles in Hot-dipped Zn-Ti Coating

引用

Journal of Iron and Steel Research International 2014年第3期21卷 396-402页

作者： Yan GUI qiao-yu XU Yan-ling GUO Guangzhou Panyu Polytechnic South China University of Technology

Hot-dip Zn-Ti galvanizing can restrain the excess growth of the coating which has better corrosion resist- ances than hot-dip pure Zn. The change rules of Г2 intermetallic compound in hot-dip Zn-Ti galvanizing is investiga ted by immersion time, titanium content and the slag test using scanning electron microscopy （SEM）, energy dis- persive spectroscopy （EDS） and X-ray diffraction （XRD）. And the mechanism of nucleation and growth of Г2 inter- metallic compound is analysed. The results show that, when adding 0. 05mass% titanium to the zinc bath, the Г2 particles form in the η layer, which nucleate by obtaining Ti atoms from the phase and Fe atoms from the dissol- ving ζ phase layer. Moreover, the more titanium is added into the bath, the more and bigger Гz particles appear. The Fz particles in the coatings grow up noticeably with prolonging the immersion time.

关键词： hot-dip galvanizing Zn-Ti alloy bath Г2 phase particle

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：