Log in

Graphical Models of Probability
Bayesian Networks, Decision Networks, and Probabilistic Relational Models
Using GP to build Bayes Net 
29th-Dec-2005 04:10 am
recently I was asked by a friend on if it would be possible to build/evolve Bayes Net with Genetic Programming. I don't know how to answer it as I only know the very basic of how Bayesnet works ... What do you think ?

One of the first questions that I raise is why needs to use GP to build BN. Can a BN model be at such a difficulty level that it requires a heuristic approach such as GP ?
29th-Dec-2005 03:44 pm (UTC)
It may make sense for some applications, but not all.

Yes, Bayes nets will almost certainly be sufficiently complex to require heavyweight approaches, particularly if assembling them from some kind of raw data.
This page was loaded Feb 24th 2017, 10:41 am GMT.