期刊文献+
共找到83,646篇文章
< 1 2 250 >
每页显示 20 50 100
K 1,3 -Free图上的弦泛圈性
1
作者 李欢 田增娴 杨卫华 《应用数学进展》 2024年第5期1994-1999,共6页
一个非诱导圈被称为弦圈,即在图中至少有一条额外的边连接圈内两个非相邻的顶点。一个阶数为n的图G, 如果G包含长度为从4到n的弦圈,则称为弦泛圈图。1991年,R.J. Faudree, R.J. Gould和T.E. Lindquester得出结论:令G是阶数为n(n≥14)的2... 一个非诱导圈被称为弦圈,即在图中至少有一条额外的边连接圈内两个非相邻的顶点。一个阶数为n的图G, 如果G包含长度为从4到n的弦圈,则称为弦泛圈图。1991年,R.J. Faudree, R.J. Gould和T.E. Lindquester得出结论:令G是阶数为n(n≥14)的2-连通,K1,3-free图。如果对于每一对不相邻的顶点x,y,有|N(x)∪N(y)|≥2n−23,则G是泛圈图。在本文中,我们扩展了这个结果,把泛圈性推广到弦泛圈性:对于任意2-连通,阶数为n(n≥34)的K1,3-free图G,如果每对不相邻顶点x,y∈V(G),满足|N(x)∪N(y)|≥2n−23,则图G是弦泛圈图。 展开更多
关键词 -free 弦圈 泛圈 弦泛圈 哈密尔顿
下载PDF
基于label-free定量蛋白质组学方法筛选沉默CHAF1B基因后心肌细胞差异表达蛋白及调控网络分析
2
作者 康彦红 顾爱琴 +1 位作者 张莹 黄帅 《首都医科大学学报》 CAS 北大核心 2024年第2期312-321,共10页
目的分析沉默染色质装配因子1亚基B(chromatin assembly factor 1 subunit B,CHAF1B)基因后心肌细胞中差异表达蛋白,预测CHAF1B基因调控网络,为寻找促进心肌细胞修复的潜在治疗靶点提供参考。方法采用转染和蛋白质印迹法筛选沉默CHAF1B... 目的分析沉默染色质装配因子1亚基B(chromatin assembly factor 1 subunit B,CHAF1B)基因后心肌细胞中差异表达蛋白,预测CHAF1B基因调控网络,为寻找促进心肌细胞修复的潜在治疗靶点提供参考。方法采用转染和蛋白质印迹法筛选沉默CHAF1B基因的有效小干扰RNA(small interfering RNA,siRNA)。应用有效siRNA沉默人源心肌AC16细胞CHAF1B基因后,采用细胞活力检测方法检测细胞活力;提取总蛋白质进行定量、还原、烷基化和胰蛋白酶裂解成肽段,利用高效液相串联质谱法鉴定肽段;搜索UniProt蛋白库筛选差异表达的蛋白质进行基因本体(Gene Ontology,GO)富集分析、京都基因与基因组百科全书(Kyoto Encyclopedia of Genes and Genomes,KEGG)通路富集和蛋白质互作网络(protein-protein interaction networks,PPI)分析。结果siRNA有效沉默CHAF1B基因后,心肌细胞存活明显受到抑制;label-free定量蛋白质组学方法鉴定结果显示,共有69个差异表达蛋白质,其中50个表达显著上调(差异倍数≥2,P<0.05),19个表达显著下调(差异倍数≤0.5,P<0.05)。GO分析显示,差异表达蛋白质主要参与大分子复合亚基体、细胞组分生物合成和组装等生物学过程,分布在细胞质和囊泡等区域,发挥蛋白质结合等分子功能。KEGG通路富集和PPI分析显示,差异表达蛋白质参与的信号通路包括蛋白酶体、氨酰tRNA生物合成、胞吞、嘧啶代谢和氨基酸生物合成等10条信号途径;表达显著上调的蛋白质如蛋白酶体亚单位A2和B7、26 S蛋白酶体调节亚单位6B和10B参与蛋白酶体途径,丝氨酸、甘氨酸、谷氨酰胺和赖氨酸tRNA合成酶介导氨酰tRNA生物合成;表达显著下调的蛋白质包括骨架相关蛋白2/3复合体亚单位3和热休克70蛋白1样参与胞吞作用,核糖核苷-二磷酸还原酶大亚基介导嘧啶代谢等通路。实时荧光定量聚合酶链式反应结果证实,转染CHAF1B siRNA后心肌细胞中合成骨架相关蛋白2/3复合体亚单位3的基因ARPC3和氨酰tRNA生物合成关键基因QARS1的mRNA水平均显著降低。结论CHAF1B为心肌细胞存活的关键蛋白质,参与调控心肌细胞的胞吞和氨基酸生物合成等多种生物学过程,参考其调控网络可帮助寻找促进心肌细胞修复的干预环节。 展开更多
关键词 label-free定量蛋白质谱 染色质装配因子1亚基B 基因敲低
下载PDF
基于一种距离相关的超高维生存数据Model-Free特征筛选
3
作者 潘莹丽 王昊宇 +1 位作者 喻佳丽 刘展 《湖北大学学报(自然科学版)》 CAS 2024年第1期122-132,共11页
随着大数据时代的来临,数据维度爆炸式增长,超高维数据的降维问题逐渐成为众多研究领域的热点话题。由于响应变量通常存在右删失,处理超高维完全数据的降维方法在右删失数据中将不再适用。本研究提出一种新的基于距离相关能有效处理超... 随着大数据时代的来临,数据维度爆炸式增长,超高维数据的降维问题逐渐成为众多研究领域的热点话题。由于响应变量通常存在右删失,处理超高维完全数据的降维方法在右删失数据中将不再适用。本研究提出一种新的基于距离相关能有效处理超高维右删失数据的特征筛选方法。首先利用距离相关系数计算每个协变量对响应变量的边际效应,建立与该系数有关的筛选指标,然后再根据事先确立的筛选准则进行特征筛选。提出的特征筛选方法不依赖任何模型结构假定,因此可以有效避免模型指定错误带来的不良后果。此外,该方法采用的距离协方差估计量是总体距离协方差的一个无偏估计,统计准确性和计算精度高。模拟和实证研究表明,提出的方法能在保留所有重要变量的前提下快速剔除与响应变量相关程度较弱的协变量,从而达到降低参数维数的目的。 展开更多
关键词 超高维数据 生存数据 距离相关 Model-free特征筛选
下载PDF
基于anchor-free的光学遥感舰船关重部位检测算法
4
作者 张冬冬 王春平 付强 《北京航空航天大学学报》 EI CAS CSCD 北大核心 2024年第4期1365-1374,共10页
针对基于深度学习的遥感舰船检测算法存在精细化程度不足、检测效率低的问题,提出一种基于anchor-free的光学遥感舰船关重部位检测算法。所提算法以全卷积的单阶段目标检测(FCOS)算法为基准,在主干网络中引入全局上下文模块,提高网络的... 针对基于深度学习的遥感舰船检测算法存在精细化程度不足、检测效率低的问题,提出一种基于anchor-free的光学遥感舰船关重部位检测算法。所提算法以全卷积的单阶段目标检测(FCOS)算法为基准,在主干网络中引入全局上下文模块,提高网络的特征表达能力;为更好地描述目标的方向性,在预测阶段构建了具有方向表征能力的回归分支;对中心度函数进行优化,使其具备方向感知和自适应能力。实验结果表明:在自建舰船关重部位数据集和HRSC2016上,所提算法的平均精度(AP)比FCOS算法有显著提升;与其他算法相比,所提算法在检测速度和检测精度上均表现优越,具有较高的检测效率。 展开更多
关键词 深度学习 遥感图像 anchor-free 舰船检测 关重部位检测 全卷积单阶段检测
下载PDF
Anchor-free目标检测算法综述
5
作者 陈恒星 刘一鸣 《机电工程技术》 2024年第8期7-12,共6页
目标检测作为计算机视觉领域的基础,其研究价值对于推动人工智能发展具有重要意义。长期以来,许多学者都致力于提升目标检测算法效率及性能的研究,无锚点(anchor-free)的目标检测深度学习算法以尺度灵活、鲁棒性强等优势,开始逐渐广泛... 目标检测作为计算机视觉领域的基础,其研究价值对于推动人工智能发展具有重要意义。长期以来,许多学者都致力于提升目标检测算法效率及性能的研究,无锚点(anchor-free)的目标检测深度学习算法以尺度灵活、鲁棒性强等优势,开始逐渐广泛应用于目标检测任务。介绍了目标检测领域中卷积神经网络和Transformer两种经典的网络架构;以核心网络架构为分类标准,分别介绍了基于卷积神经网络和基于Transformer的anchor-free目标检测深度学习算法,总结了这些算法的改进点和优缺点,并对该方向的未来发展及应用做出展望。 展开更多
关键词 目标检测 anchor-free 卷积神经网络 TRANSFORMER
下载PDF
Mechanical and Electrical Properties of Some Sn-Zn Based Lead-Free Quinary Alloys
6
作者 Shihab Uddin Md. Abdul Gafur +1 位作者 Suraya Sabrin Soshi Mohammad Obaidur Rahman 《Materials Sciences and Applications》 2024年第7期213-227,共15页
Although there are many lead-free soldering alloys on the market, none of them have ideal qualities. The researchers are combining binary alloys with a variety of additional materials to create the soldering alloys’ ... Although there are many lead-free soldering alloys on the market, none of them have ideal qualities. The researchers are combining binary alloys with a variety of additional materials to create the soldering alloys’ features. The eutectic Sn-9Zn alloy is among them. This paper investigated the mechanical and electrical properties of Sn-9Zn-x (Ag, Cu, Sb);{x = 0.2, 0.4, and 0.6} lead-free solder alloys. The mechanical properties such as elastic modulus, ultimate tensile strength (UTS), yield strength (YS), and ductility were examined at the strain rates in a range from 4.17 10−3 s−1 to 208.5 10−3 s−1 at room temperature. It is found that increasing the content of the alloying elements and strain rate increases the elastic modulus, ultimate tensile strength, and yield strength while the ductility decreases. The electrical conductivity of the alloys is found to be a little smaller than that of the Sn-9Zn eutectic alloy. 展开更多
关键词 Lead-free Solder Strain Rate Ultimate Tensile Strength DUCTILITY Electrical Conductivity
下载PDF
基于Anchor-free的孪生网络目标跟踪算法
7
作者 苑侗侗 杨文柱 +1 位作者 李前 王玉霞 《计算机应用与软件》 北大核心 2023年第12期176-183,216,共9页
基于anchor-free的目标预测方法相较于anchor-based方法速度更快,故设计一种基于anchor-free的孪生网络目标跟踪算法(AFSN)。通过对特征图、预测结果进行双重融合来提升跟踪效果,由深至浅对特征图进行堆叠融合,利用多层特征图进行目标预... 基于anchor-free的目标预测方法相较于anchor-based方法速度更快,故设计一种基于anchor-free的孪生网络目标跟踪算法(AFSN)。通过对特征图、预测结果进行双重融合来提升跟踪效果,由深至浅对特征图进行堆叠融合,利用多层特征图进行目标预测,融合多个预测结果来稳定跟踪效果。采用anchor-free的目标预测方法,直接在像素点上进行目标类别的预测和边界框回归,避免了需设计大量锚点包围盒的问题。在GOT-10K数据集上,该算法的平均重叠率(AO)和成功率(SR_(0.75))相较于SiamRPN++算法提高了4.9和9.9百分点,算法处理速度可达每秒37帧。 展开更多
关键词 目标跟踪 孪生网络 anchor-free 双重融合
下载PDF
基于Label-free技术的高州油茶铝胁迫蛋白质组学研究
8
作者 程俊森 王溢 +4 位作者 李永泉 魏尚霖 李超楠 姜维 黄润生 《中南林业科技大学学报》 CAS CSCD 北大核心 2023年第8期169-181,共13页
【目的】南方地区拥有大面积的酸性红壤,导致土壤中铝元素多以离子的形式溶出,严重抑制油茶的生长及高产。研究采用Label-free技术探究高州油茶在铝胁迫下的蛋白质组学响应,为揭示高州油茶应答铝胁迫的分子响应机制奠定了理论基础。【... 【目的】南方地区拥有大面积的酸性红壤,导致土壤中铝元素多以离子的形式溶出,严重抑制油茶的生长及高产。研究采用Label-free技术探究高州油茶在铝胁迫下的蛋白质组学响应,为揭示高州油茶应答铝胁迫的分子响应机制奠定了理论基础。【方法】以高州油茶两年生实生苗为试验材料,测定不同浓度铝处理条件下高州油茶叶片部分代谢指标及抗氧化生理相关指标,同时比较对照组(CK)和4 mmol/L铝处理组(GZ4)差异蛋白质的表达。【结果】生理试验结果表明,不同浓度铝胁迫下油茶幼苗都会发生氧化胁迫导致细胞活性氧(ROS)积累,2 mmol/L铝浓度下油茶叶片通过增加渗透调节物质含量和增强多种抗氧化酶活性以清除细胞活性氧,使细胞免受氧化胁迫的损害;而4 mmol/L铝浓度下叶片丙二醛(MDA)含量和脯氨酸(PRO)含量明显升高,积累的活性氧超出了自身所能清除的范围,叶片可溶性糖含量和可溶性蛋白含量显著减少,抗氧化酶活性均有不同程度减弱,油茶正常的生理功能受到抑制。蛋白质组学数据表明,CK组和GZ4组共鉴定到4282个蛋白质,与CK组相比,GZ4处理组有207个蛋白质的表达丰度显著增加,129个蛋白质的表达丰度显著降低。差异表达蛋白质的通路富集和蛋白质互作网络分析结果表明,4 mmol/L铝胁迫下高州油茶主要通过提高谷胱甘肽代谢,增强细胞清除活性氧和自由基负离子的能力缓解铝胁迫造成的氧化损伤;4 mmol/L铝浓度下高州油茶蛋白质合成和光合作用受到明显抑制,抗氧化酶活性减弱可能与细胞供能能力不足有关。【结论】铝胁迫下高州油茶主要通过能量及碳水化合物代谢、氨基酸合成及代谢、生物碱合成、蛋白质加工、卟啉和叶绿素代谢等代谢途径抵抗铝毒。 展开更多
关键词 高州油茶 铝胁迫 蛋白质组学 LABEL-free
下载PDF
{nest,gap}-free图的边理想正则度的研究
9
作者 杨娟 刘阿明 《海南大学学报(自然科学版)》 CAS 2023年第2期115-120,共6页
gap-free图是指不含gap作为导出子图的图,其中gap是顶点集为{a,b,u,v}和边集为{ab,uv}的图.证明了所有的nest-free且gap-free图的边理想正则度reg I((G))是小于等于3的.定义了n-gap-free图,并刻画了一些n-gap-free图的边理想的正则度.
关键词 gap-free 图的边理想 Castelnuovo-Mumford正则度 自由预解式
下载PDF
Effect of Strain Rate on Tensile Behavior of Sn-9Zn-xAg-ySb;{(x, y) = (0.2, 0.6), (0.2, 0.8), (0.6, 0.2), (0.8, 0.2)} Lead-Free Solder Alloys
10
作者 Shihab Uddin Md. Abdul Gafur Mohammad Obaidur Rahman 《Materials Sciences and Applications》 CAS 2023年第4期273-283,共11页
The tensile properties of Sn-9Zn-xAg-ySb;{(x, y) = (0.2, 0.6), (0.2, 0.8), (0.6, 0.2), (0.8, 0.2)} lead-free solders were investigated. All the test samples were annealed at 150°C for 1 hour. The tests are carrie... The tensile properties of Sn-9Zn-xAg-ySb;{(x, y) = (0.2, 0.6), (0.2, 0.8), (0.6, 0.2), (0.8, 0.2)} lead-free solders were investigated. All the test samples were annealed at 150°C for 1 hour. The tests are carried out at room temperature at the strain rate of 4.17 × 10<sup>-3</sup> s<sup>-1</sup>, 20.85 × 10<sup>-3</sup> s<sup>-1</sup>, and 208.5 × 10<sup>-3</sup> s<sup>-1</sup>. It is seen that the tensile strength increases and the ductility decrease with increasing the strain rate over the investigated range. From the strain rate change test results, the strain sensitivity values are found in the range of 0.0831 to 0.1455 due to the addition of different alloying elements. 展开更多
关键词 Lead-free Solder Strain Rate Strain Sensitivity DUCTILITY Tensile Properties
下载PDF
Model-Free Feature Screening via Maximal Information Coefficient (MIC) for Ultrahigh-Dimensional Multiclass Classification
11
作者 Tingting Chen Guangming Deng 《Open Journal of Statistics》 2023年第6期917-940,共24页
It is common for datasets to contain both categorical and continuous variables. However, many feature screening methods designed for high-dimensional classification assume that the variables are continuous. This limit... It is common for datasets to contain both categorical and continuous variables. However, many feature screening methods designed for high-dimensional classification assume that the variables are continuous. This limits the applicability of existing methods in handling this complex scenario. To address this issue, we propose a model-free feature screening approach for ultra-high-dimensional multi-classification that can handle both categorical and continuous variables. Our proposed feature screening method utilizes the Maximal Information Coefficient to assess the predictive power of the variables. By satisfying certain regularity conditions, we have proven that our screening procedure possesses the sure screening property and ranking consistency properties. To validate the effectiveness of our approach, we conduct simulation studies and provide real data analysis examples to demonstrate its performance in finite samples. In summary, our proposed method offers a solution for effectively screening features in ultra-high-dimensional datasets with a mixture of categorical and continuous covariates. 展开更多
关键词 Ultrahigh-Dimensional Feature Screening MODEL-free Maximal Information Coefficient (MIC) Multiclass Classification
下载PDF
Effect of Temperature on Frost-Free Days and Length of Crop Growing Season across Southern Ontario
12
作者 Ramesh Pall Rudra Rituraj Shukla +2 位作者 Trevor Dickinson Pradeep Kumar Goel Jaskaran Dhiman 《American Journal of Climate Change》 2023年第4期700-718,共19页
Climate change has an impact on various climatic variables. In this study our focus is mainly on temperature characteristics of climate parameter. In temperate and humid regions like southern Ontario, the effect of cl... Climate change has an impact on various climatic variables. In this study our focus is mainly on temperature characteristics of climate parameter. In temperate and humid regions like southern Ontario, the effect of climate change on Frost-free days in winter is distinctive. The average annual temperature is going upward but the extreme increase is in the winter temperature. Winter average temperature is going up by about 2˚C. However, extreme daily minimum temperature is going up by more than 3˚C. This climate effect has a great impact on the nature of precipitation and length of frost-free days. The snowfall over winter months is decreasing and the rainfall is increasing. However, the number of frost-free days during late fall months, early winter months, late winter months and early spring months are increasing. This result reveals an increase in length of the growing season. This research focuses on the effect of change in climatic variables on Frost-free days in Southern Ontario. Therefore, special attention should be given to the effect of change in climate Frost-free conditions on length of crop growing in winter season for potential investigation. 展开更多
关键词 Climate Change Minimum Temperature Frost-free Days SNOWFALL Crop Growing Season Southern Ontario
下载PDF
Model-Free Ultra-High-Dimensional Feature Screening for Multi-Classified Response Data Based on Weighted Jensen-Shannon Divergence
13
作者 Qingqing Jiang Guangming Deng 《Open Journal of Statistics》 2023年第6期822-849,共28页
In ultra-high-dimensional data, it is common for the response variable to be multi-classified. Therefore, this paper proposes a model-free screening method for variables whose response variable is multi-classified fro... In ultra-high-dimensional data, it is common for the response variable to be multi-classified. Therefore, this paper proposes a model-free screening method for variables whose response variable is multi-classified from the point of view of introducing Jensen-Shannon divergence to measure the importance of covariates. The idea of the method is to calculate the Jensen-Shannon divergence between the conditional probability distribution of the covariates on a given response variable and the unconditional probability distribution of the covariates, and then use the probabilities of the response variables as weights to calculate the weighted Jensen-Shannon divergence, where a larger weighted Jensen-Shannon divergence means that the covariates are more important. Additionally, we also investigated an adapted version of the method, which is to measure the relationship between the covariates and the response variable using the weighted Jensen-Shannon divergence adjusted by the logarithmic factor of the number of categories when the number of categories in each covariate varies. Then, through both theoretical and simulation experiments, it was demonstrated that the proposed methods have sure screening and ranking consistency properties. Finally, the results from simulation and real-dataset experiments show that in feature screening, the proposed methods investigated are robust in performance and faster in computational speed compared with an existing method. 展开更多
关键词 Ultra-High-Dimensional Multi-Classified Weighted Jensen-Shannon Divergence MODEL-free Feature Screening
下载PDF
复杂系统的Scale-free性及其宏观调控问题 被引量:2
14
作者 戴冠中 王林 覃森 《科技导报》 CAS CSCD 2006年第5期11-15,共5页
系统研究了复杂系统的Scale-free问题,包括复杂系统的Scale-free性、幂律分布的分析、Scale-free性的形成机制、复杂系统的Scale-free现象,以及复杂系统Scale-free性的宏观调控等问题。研究结果强调:(1)从广义上说,Scale-free性是描述... 系统研究了复杂系统的Scale-free问题,包括复杂系统的Scale-free性、幂律分布的分析、Scale-free性的形成机制、复杂系统的Scale-free现象,以及复杂系统Scale-free性的宏观调控等问题。研究结果强调:(1)从广义上说,Scale-free性是描述大量复杂系统整体上严重不均匀分布的一种内在性质;(2)证明了只有当幂律指数在2~3之间时,用幂律分布所描述的大量复杂系统才具有严重的不均匀分布特性;(3)对于国计民生有重大影响的复杂系统,提出了Scale-free性的宏观调控的原则方法。 展开更多
关键词 复杂系统 复杂网络 Scale-free 幂律分布 Scale-free现象 宏观调控
下载PDF
Revolutionizing Non-Invasive Biomarker Discoveries: The Power of Methylation Screening Analysis in Cell-Free DNA Liquid Biopsy
15
作者 Min Seob Lee Na Young Min +2 位作者 Hyuk Jung Kwon Yonjung Kim Isaac Kise Lee 《Open Journal of Genetics》 CAS 2023年第1期48-74,共27页
Epigenetic changes of DNA, including methylation, have long been recognized as key indicators of various diseases, including aging, cancer, and neurological disorders. Biomarker discoveries based on distinct methylati... Epigenetic changes of DNA, including methylation, have long been recognized as key indicators of various diseases, including aging, cancer, and neurological disorders. Biomarker discoveries based on distinct methylation patterns for both hypermethylation and hypomethylation lead the way in discovery of novel diagnosis and treatment targets. Many different approaches are present to detect the level of methylation in whole genome (whole genome bisulfite sequencing, microarray) as well as at specific loci (methylation specific PCR). Cell-free DNA (cf-DNA) found in body fluids like blood provides information about DNA methylation and serves as a less invasive approach for genetic screening. Cell-free DNA and methylation screening technologies, when combined, have the potential to transform the way we approach genetic screening and personalized therapy. These technologies can help enhance disease diagnostic accuracy and inform the development of targeted therapeutics by providing a non-invasive way for acquiring genomic information and identifying disease-associated methylation patterns. We highlight the clinical benefits of using cell-free DNA (cf-DNA) liquid biopsy analysis and available methylation screening technologies that have been crucial in identifying biomarkers for disease from patients using a non-invasive way. Powering such biomarker discoveries are various methods of cf-DNA methylation analysis such as Bisulfite Sequencing and most recently, Methylation-Specific Restriction Enzyme (MSRE-seq) Analysis, paving the way for novel epigenetic biomarker discoveries for more robust diagnosis such as early disease detection, prognosis, monitoring of disease progression and treatment response as well as discovery of novel drug targets. 展开更多
关键词 Epigenetics Biomarkers Cell-free DNA (cf-DNA) METHYLATION Liquid Biopsy Drug Target Methylation-Specific Restriction Enzyme (MSRE) Cancer Epigenetic Drugs HYPERMETHYLATION HYPOMETHYLATION
下载PDF
基于双生成器网络的Data-Free知识蒸馏 被引量:1
16
作者 张晶 鞠佳良 任永功 《计算机研究与发展》 EI CSCD 北大核心 2023年第7期1615-1627,共13页
知识蒸馏(knowledge distillation,KD)通过最大化近似输出分布使“教师网络”指导“学生网络”充分训练,成为大规模深度网络近端迁移、部署及应用的重要技术.然而,隐私保护意识增强与传输问题加剧使网络训练数据难以获取.如何在Data-Fre... 知识蒸馏(knowledge distillation,KD)通过最大化近似输出分布使“教师网络”指导“学生网络”充分训练,成为大规模深度网络近端迁移、部署及应用的重要技术.然而,隐私保护意识增强与传输问题加剧使网络训练数据难以获取.如何在Data-Free的自由环境下,保证压缩网络准确率成为重要的研究方向.Data-Free学生网络学习(data-free learning of student networks,DAFL)模型,建立“教师”端生成器获得与预训练网络分布近似的伪数据集,通过知识蒸馏训练“学生网络”.然而,该框架中生成器构建及优化仍存在2个问题:1)过度信任“教师网络”对缺失真实标签伪样本的判别结果,同时,“教师网络”与“学生网络”优化目标不同,使“学生网络”难以获得准确、一致的优化信息;2)仅依赖于“教师网络”训练损失,导致数据特征多样性缺失,降低“学生网络”泛化性.针对这2个问题,提出双生成器网络架构DG-DAFL(double generators-DAFL),分别建立“教师”与“学生”端生成器并同时优化,实现网络任务与优化目标一致,提升“学生网络”判别性能.进一步,增加双生成器样本分布差异损失,利用“教师网络”潜在分布先验信息优化生成器,保证“学生网络”识别准确率并提升泛化性.实验结果表明,该方法在Data-Free环境中获得了更为有效且更鲁棒的知识蒸馏效果.DG-DAFL方法代码及模型已开源:https://github.com/LNNU-computer-research-526/DG-DAFL.git. 展开更多
关键词 深度神经网络 知识蒸馏 无数据环境知识蒸馏 对抗生成网络 生成器
下载PDF
基于Anchor-free的图像全景分割算法研究
17
作者 陈颖 秦怡芳 张艳硕 《计算机应用与软件》 北大核心 2023年第10期255-260,290,共7页
对于实例分割子任务,大多数框架都采用基于anchor的策略,如著名的Mask R-CNN。但anchor依赖于人为预先设置,且由于anchor和RPN的存在,网络计算量往往很大。设计一个自定义的U-Net骨干网络和一个立足于极坐标系对实例分割掩膜边缘进行数... 对于实例分割子任务,大多数框架都采用基于anchor的策略,如著名的Mask R-CNN。但anchor依赖于人为预先设置,且由于anchor和RPN的存在,网络计算量往往很大。设计一个自定义的U-Net骨干网络和一个立足于极坐标系对实例分割掩膜边缘进行数学建模的anchor-free的实例分割网络,将二者集成到全景分割架构中去。该全景分割网络摆脱人工预设的anchor,实现轻量级,U-Net骨架网相比FCN网干网在精度相差不多的情况下参数量下降35%,更适合低开销实时应用场景。 展开更多
关键词 全景分割 实例分割 无锚框 极坐标系 自定义U-Net
下载PDF
Calculation of Probability of Cloud-Free Lines-of-Sightat Given Heights in Foshan, China 被引量:1
18
作者 李昀英 孙立潭 《Advances in Atmospheric Sciences》 SCIE CAS CSCD 1999年第2期297-303,共7页
Based on Lund and Shanklin’s work (1972), methods of calculating Probability of Cloud-Free Lines-of-Sight (PCFLOS), Persistence Probability of Cloud-Free Lines-of-Sight (PPCFLOS) and Recurrence Probability of Cloud-f... Based on Lund and Shanklin’s work (1972), methods of calculating Probability of Cloud-Free Lines-of-Sight (PCFLOS), Persistence Probability of Cloud-Free Lines-of-Sight (PPCFLOS) and Recurrence Probability of Cloud-free Lines-of-Sight (RPCFLOS) at given heights are presented. PCFLOS, PPCFLOS and RPCFLOS are calculated in Foshan, China by conventional observation data from 1961 to 1990. The conclusions are: (1) The higher the elevations, the smaller the PCFLOS and the larger the view angles, the larger the PCFLOS. (2) PPCFLOS and RPCFLOS decrease with the increase of elevation and the delay of time. (3) RPCFLOS is always equal to or larger than PPCFLOS at lag times. 展开更多
关键词 Probability of cloud-free lines-of-sight Persistence probability of cloud-free lines-of-sight Recurrence probability of cloud-free lines-of-sight Cloud climate
下载PDF
基于Label-free技术的非小细胞肺癌蛋白质差异研究
19
作者 魏文海 李兴芳 +4 位作者 张新迪 周立文 赵琼 李静芸 牛慧敏 《内蒙古医学杂志》 2023年第9期1025-1028,F0002,F0003,共6页
目的 基于非小细胞肺癌(non-small-cell lung cancer,NSCLC)及与其配对的非癌性邻近组织(non-cancerous adjacent tissues,NATs)的蛋白质组学分析,寻找NSCLC诊断相关蛋白标志物。方法 应用非标记定量蛋白组学(LFP)技术,以NSCLC患者配对... 目的 基于非小细胞肺癌(non-small-cell lung cancer,NSCLC)及与其配对的非癌性邻近组织(non-cancerous adjacent tissues,NATs)的蛋白质组学分析,寻找NSCLC诊断相关蛋白标志物。方法 应用非标记定量蛋白组学(LFP)技术,以NSCLC患者配对的NATs为对照,对5例NSCLC患者进行癌组织蛋白质组学分析。以差异倍数>1.5(上调)或<0.67(下调)且P<0.05为标准筛选差异蛋白。应用String网站和R语言对差异蛋白进行基因本体论(gene ontology,GO)分析、京都基因和基因组百科全书(Kyoto Encyclopedia of Genes and Genomes,KEGG)分析,为进一步研究提供良好基础。结果 本研究发现差异蛋白共644个,其中339个蛋白表达上调,305个蛋白表达下调。PCA结果显示,差异蛋白可明显区分NSCLC与NATs。GO分析结果表明,差异蛋白大多以细胞外泌体的形式存在且主要富集于调节胞外分泌、胞外分泌、骨髓细胞激活参与免疫反应等的生物学过程以及钙粘着蛋白绑定、细胞外基质绑定等的分子功能。KEGG分析结果显示,差异蛋白主要富集在抗坏血酸和醛酸代谢、组氨酸代谢、蛋白质消化吸收、丙酮酸代谢等通路(P<0.05)。结论 NSCLC与NATs存在明显差异,可为发现NSCLC新型标志物提供线索。 展开更多
关键词 非小细胞肺癌 非标记定量蛋白质组学 新型蛋白标志物
下载PDF
关于f-free子集
20
作者 巩长忠 《辽宁工学院学报》 1996年第4期71-73,共3页
在整数集或有理数域Q上的向量空间的有限子集上建立了几个函数,证明了这些函数在整数集合内取得它们的最小值。作为它们的一个推论,证明了特征为零的任何非零元素的有限集合都包含一个sum-free子集B,|B|>13|A|.
关键词 f-free子集 sum-free 阿贝尔群 向量空间
下载PDF
上一页 1 2 250 下一页 到第
使用帮助 返回顶部