em algorithm造句

• Research on unsupervised chinese segmentation based on em algorithm
  一种改进的汉语分词算法
• Studies on correct convergence of the em algorithm for gaussian mixtures
  算法正确收敛性的探讨
• Study of tracking maneuvering targets in clutter based on the em algorithm
  性能优化的跟踪门算法
• We first introduce the em algorithm and its convergence properties
  首先，我们引入了em算法，讨论了它的一般收敛性质。
• We also give a variation of the em algorithm for gaussian mixtures
  进一步给出了高斯混合体em算法的变形及其收敛性质。
• Correction of control point slope based on em algorithm and shading of single sar image
  算法和单幅雷达图像阴影的控制点坡度校正
• Image segmentation ; textured image ; gaussian mixture models ; multiresolution wavelet ; em algorithms
  图像分割纹理图像高斯混合模型多分辨率小波em算法
• 18 dempster a , laird n , rubin d . maximum likelihood from incomplete data via the em algorithm
  进一步，数据降维的结果可通过与两个空间相关度量的最大似然解得到。
• In this paper , we present a theoretical analysis on the correct convergence of the em algorithm for gaussian mixtures
  本文对高斯混合体em算法的正确收敛性问题进行了理论研究。
• However , there has been still an important but unsolved problem : whether the em algorithm can converge to the correct solution
  然而， em算法能否收敛到正确解一直是一个困难的问题。
• It uses gassian mixture model to represent particles and adopts em algorithm to refit particles after correction step at each time
  该算法使用混合高斯模型表示粒子，在每个时刻的修正步骤之后，采用em算法对粒子进行重新拟合。
• Journal of machine learning research , 2004 , 5 : 913 - 939 . 16 dempster a p , laird n m , rubin d b . maximum likelihood from incomplete data via the em algorithm
  这个一般性的法则揭示了多示例学习与监督学习之间的联系，为多示例学习算法的设计提供了一种通用的途径。
• Since em algorithm has certain good convergence properties and can be widely applied , it has attracted many researchers from statistics , other applied science and engineering
  由于具有很好的收敛性质和广泛的应用背景， em算法一直受到统计学和其它应用科学及工程界的广泛重视。
• We further obtain and prove the condition of the correct convergence of the em algorithm for gaussian mixtures . finally , the theoretical results have been demonstrated by the simulation results
  进一步，我们得到了高斯混合密度的em算法正确收敛性条件，并对此进行了理论证明和数值验证。
• But in practical applications and experiments , we often find that em algorithm for gaussian mixtures can converge correctly when the overlap measure of component densities in the mixtures becomes small
  但在实际应用与实验中，我们经常发现当重叠度较小时，概率混合体em算法往往收敛到正确解。
• Actually , it has been found that the convergence rate of the em algorithm for gaussian mixtures is related to the overlap measure of gaussians in the mixture in certain situation
  近年来，对于高斯混合体em算法的收敛性研究有了新的进展。人们发现算法的收敛速度与高斯混合体各分量之间的重叠度有关。
• Two steps were included : the first step was to label those documents containing latent class variables by blsa ; the second step was to label the rest by naive bayesian model with em algorithm
  本算法不需要对大量训练样本的类别标注，只需提供相应的类主题变量，从而提高了wsdl文档分类的自动性和效率。
• That is , the convergence speed of the em algorithm becomes much fast as the overlap measure of gaussians in the mixture tends to zero , which conforms to simulation results and practical applications
  当这一重叠度趋向于零时，算法的收敛速度渐进地趋向于超线性，变成为一个快速算法。这与实际应用是相符合的。
• This paper researched a learning algorithm of bayesian networks in incomplete data which algorithm combined niche genetic algorithm with em algorithm . then the experiment shows the algorithm is valid
  在数据缺失的情况下讨论一种贝叶斯网络的结构学习算法。该算法结合了小生境遗传算法和em算法,最后通过试验说明了该算法的有效性。
• We then prove that the em algorithm becomes a compact map in certain neighborhood of a consistent solution when a measure of the average overlap of gaussians in the mixtures is small enough and the sample size is large enough
  我们证明了高斯混合体em算法在混合密度的重叠度很小时在其样本真解相一致的解的一个邻域内是一个压缩映射。