Directed Graphical Statistical Model Family
(Redirected from Directed graphical models)
Jump to navigation
Jump to search
A Directed Graphical Statistical Model Family is a graphical model family that uses directed edges.
- Context:
- It can be associated with a Directed Probabilistic Network.
- It can range from being an Directed Conditional Graphical Model Family to being an Directed Joint Graphical Model Family.
- …
- Example(s):
- Counter-Example(s):
- See: Statistical Model.
References
2014
- http://factorie.cs.umass.edu/usersguide/UsersGuide030Overview.html
- Directed graphical models (also known as Bayesian networks) represent a joint distribution over random variables by a product of normalized conditional probability distributions (one for each "child" variable, conditioned on the other "parent" variables that are connected by an incoming directed edge). They are convenient generative models when variable values can be generated by an ordered iteration of the graph nodes from "parent" variables to "child" variables.