kensun0的专栏

关于Gibbs sampling， 首先看一下Wiki上的解释：Gibbs sampling or Gibbs sampler is analgorithm to generate a sequence of samples from the joint probability distribution of two or morerandom variables. The purpose of

梯度：函数f(x)在某一点的梯度指向最大化f(x)的方向目标：f(x)=E(x)^2=0;牛顿法：X(i+1)=X(i)-f(x)/f`(x);梯度下降法：X(i+1)=X(i)-r*f`(x);随机梯度下降法：X(i+1)=X(i)-r*f`(xi);共轭梯度法