动态模糊极大似然估计算法

Maximum Likelihood Estimation Algorithm of Dynamic Fuzzy

在机器学习中,一个广泛的应用是对模型的参数进行估计,即极大似然估计(MLE),EM算法是根据点估计中的MLE改进的一种迭代算法,是求极大似然估计的一种强有力的工具,但它收敛速度较慢,于是引入α-EM算法,克服了EM算法的缺陷。由于学习的过程中可能存在着大量的缺失数据及其动态模糊性,给出基于不完全数据的动态模糊极大似然估计算法并给出实例验证。

A wide application of machine learning is maximum likelihood estimation（MLE）. EM algorithm is an efficient tool to deal with the maximum likelihood estimation, but its convergence rate is not high. Because of many incomplete data existing in machine learning, introduces the α-EM algorithm, proposes the dynamic fuzzy maximum likelihood estimation algorithm based on α-EM and incomplete data, and gives a validated example.