Probabilistic Graphical Models 1: Representation-Week1-reasoning-patterns

课程链接:https://www.coursera.org/learn/probabilistic-graphical-models/lecture/KMjHs/reasoning-patterns

原CPD如下:

计算:

p(d1|g3) = p(d1, g3) / p(g3)

p(d1, g3) = p(g3, d1) = p(g3|d1, i0) * p(d1, i0) + p(g3|d1, i1) * p(d1, i1)

            = 0.7*0.4*0.7 + 0.2*0.4*0.3  约等于0.22

p(g3) =  p(g3|d1, i0) * p(d1, i0) + p(g3|d1, i1) * p(d1, i1) + 

      p(g3|d0, i0) * p(d0, i0) + p(g3|d0, i1) * p(d0, i1)

   =0.7*0.4*0.7 + 0.2*0.4*0.3 +

    0.3*0.6*0.7 + 0.02*0.68*0.3

    约等于0.35

p(d1|g3) =0.22/0.35 约等于0.6286

 

p(i1|g3) =  p(i1, g3) / p(g3)

p(i1, g3) = p(g3| i1, d0) * p(i1, d0) + p(g3| i1, d1) * p(i1, d1)

      = 0.02 * 0.3 * 0.6 + 0.2 * 0.3 * 0.4 

      = 0.0276

p(i1|g3) = 0.0276/0.35 = 0.079

 

p(i1 | g3, d1) = p(i1, g3, d1) / p(g3, d1)

p(i1, g3, d1)= p(g3| i1, d1) * p(i1, d1) = 0.2 * 0.4 * 0.3

p(i1 | g3, d1) =0.2 * 0.4 * 0.3/0.22 = 0.1091

posted on 2017-01-30 16:16  karis  阅读(261)  评论(0编辑  收藏  举报

导航