Quantcast
Channel: Machine Learning
Viewing all articles
Browse latest Browse all 62709

What is an example of a distribution of which you can only access an un-normalized version?

$
0
0

In the context of a lot of MCMC methods for sampling you seem to need a sampler candidate distribution Q and an un-normalized version of your target distribution P. (Where you are trying to draw samples from P) I'm wondering how this P can arise without having access to P in the context of Bayesian Inference?

submitted by nickponline
[link] [5 comments]

Viewing all articles
Browse latest Browse all 62709

Trending Articles