EM算法看一次忘一次，干脆整理一下好了。
一. 最大似然算法
基本思想：從已經(jīng)觀察到有限個(gè)samples推測出該群體最合理的distribution.
max \ Lik\left(D,z | \theta \right) in order to find the best \theta. theta is distribution parameters( means or var)
二. EM算法
基本思想：從已經(jīng)觀察到的有限個(gè)samples和一些未知的samples推測出該群體最合理的distribution.
max \ Lik\left(D,z | \theta \right) in order to find the best \theta. z are hidden variables.
具體操作
E step : fix \theta_k. max Likelihood function with respect to hidden variable z_k.
M step : fix z_k obtained from previous step, max Likelihood function with respect to the parameter \theta_{k+1}.
直覺
坐標(biāo)上升法
推導(dǎo)
Reference
http://www.cmlab.csie.ntu.edu.tw/~cyy/learning/tutorials/EM.pdf
