View Single Post
Old 01-05-2017, 06:34 PM
tushar tushar is offline
Junior Member
Join Date: Jan 2017
Posts: 2
Default Clarification of Conditional Probability Interpretation

Let's say x is determined by a normal distribution and p(y|x) is a sigmoid function if y = 1 and the sigmoid reflected along the vertical axis if y = -1.

p(y|x) = {
f(x) if y = 1
1 - f(x) if y = -1

Then p(x, y) = p(y|x) * p(x)

So for the case y = 1, does it make sense to say that p(x, y) is just the multiple of the sigmoid function and the normal distribution?
Reply With Quote