后验概率

The linguistic knowledge of Putonghua pronunciation was effectively introduced into the calculation of HMM based log posterior probability.

新算法在隐马尔科夫模型的对数后验概率算法基础上，引入普通话发音的语言学知识。

Bayesian method is used to calculate the posterior probability density function of the parameters for the model, and then Monte Carlo method is used to sample the model for getting the estimated values of the parameters.

采用贝叶斯方法计算出模型参数的后验概率密度函数，通过蒙特卡罗方法时其进行采样来获得参数的佑计值。

The posterior probability of each node in a hierarchical dynamic Bayesian network is inferred based on particle filter in order to recognize situation events in real time.

采用Particle Filter粒子滤波方法对模型中各节点状态的后验概率进行实时估计，以对场景事件进行实时识别。

Meanwhile, for the characteristics of plant disease spot, segmentation model was created by utilizing the local correlated characteristic of Markov random field and linear stationary auto-regression model formed by Gauss distribution, and self-adaptive segmentation of plant disease spot was realized through optimizing the model with improved ant colony algorithm and maximum a posteriori probability estimation of Gauss-Markov random field.

同时，利用Markov随机场的局部相关特性并结合Causs分布组成线性平稳自回归模型，针对植物病斑特征建立分割模型。最后，采用改进型蚁群算法对其进行优化，并结合Gauss-Markov随机场最大后验概率估计，实现对植物病斑的自适应分割。

Using it we can approximately obtain the expectation of any function,using the certain quantity the random sample particle to express the posterior density of random variable in the model,and can be used in the any non-linear random models.

粒子滤波算法是一种适用于非线性非正态约束的基于模拟的统计滤波算法，可以近似得到任意函数的数学期望，利用一定数量的随机样本粒子来表示模型中随机变量的后验概率分布，并且能应用于任意非线性随机模型。

In position based matching method, syllable lattice is explained as a sequence of some competition sets and then position specific posterior probability is calculated for all candidates.

在采用位置匹配机制的检索方法中，将音节Lattice解释为具有特定位置标号的若干竞争集的级联，给出了相应的搜索匹配方法，以及匹配路径处于特定位置的后验概率值的计算方法。

The fast MH mutation moves the particles so that their stationary distribution is a target posterior density.

快速MH变异能对粒子进行移动，使得粒子的稳定分布为目标的后验概率密度分布。

Based on Bellman stochastic nonlinear dynamic programming, we develop an optimal control algorithm for the discrete stochastic system with conditional Markov structure.

基于Bellman随机非线性动态规划法，提出了具有条件马尔科夫跳变结构的离散随机系统的最优控制方法，应用随机变结构系统的性质对最优控制算法进行了简化处理，并将后验概率密度函数用条件高斯函数来逼近，针对一类具有条件马尔科夫跳变结构的线性离散随机系统，给出了其逼近最优控制算法。

To lower power consumption in Turbo decoding , a new algorithm based on trellis combination was proposed.

针对Turbo 码译码器功耗大的问题，改进了传统的最大后验概率译码算法，提出了一种基于网格图合并思想的低功耗Turbo 码译码算法。

The first one is based on minimum cross entropy with the hypothesis of uniform or normal probability distribution.

这些算法与传统算法的不同主要在于：(1)基于规范的理论框架和一般性的分割准则，对问题的依赖性小；(2)采用贝叶斯后验概率计算交叉熵中的两种分布。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。