CPT计算

The engine.
Post Reply
Gary
Posts: 40
Joined: Thu Nov 17, 2016 2:38 am

CPT计算

Post by Gary »

Hello
I would like to know, in the case of each parent node independent of each other, if the father is more nodes, how to quickly calculate CPT?
Such as f(A|B)=0.5,f(A|c)=0.3,f(A|B,C) cpt is 0.15 0.85 0.5 0.5 0.3 0.7 ,But if there are more father nodes,and we know the nodes rate, how to quickly calculate CPT?
marek [BayesFusion]
Site Admin
Posts: 430
Joined: Tue Dec 11, 2007 4:24 pm

Re: CPT计算

Post by marek [BayesFusion] »

Gary,

I'm not sure I understand your question if I take it for the face value. Each column in a CPT is in general independent of any other column, so your question does not have a good answer in general.

However, your question may be interpreted differently: "Is there any structure in the CPT that you can explore to create it from your understanding of the interactions among the parents to produce the probability distribution over the states of the child node?" This is a great question of great practical importance. A node and its parents can be viewed as describing a unique (causal) mechanism acting in the system. One can imagine it as an equation describing the interaction of the parent nodes producing the child node. Very often we do have an idea how this interaction looks like, so this question has plenty of answers.

I suggest that you look at the idea of canonical models (there is a section on them in GeNIe manual). Also, please look at continuous nodes and the discretization feature/tab -- it offers a simple way of generating CPTs from equations.

Finally, your question seems to be more appropriate for GeNIe board rather than SMILE board.
I hope this helps.

Marek
Post Reply