Undirected Conditional Graphical Model Family
Jump to navigation
Jump to search
An Undirected Conditional Graphical Statical Model Family is a conditional probability graphical family that specifies a family of valid undirected conditional graphical networks.
- AKA: Undirected Conditional Graphical Statistical Metamodel, Undirected Graphical Models, Markov Networks, Markov Random Fields, MRFs.
- Context:
- It can be associated with an Undirected Conditional Probability Network.
- Example(s):
- Counter-Example(s):
- See: Directed Conditional Graphical Metamodel
References
2012
- (Barber, 2012) ⇒ David Barber. (2012). "Bayesian Reasoning and Machine Learning." In: Cambridge University Press. ISBN: 0521518148, 9780521518147.
- QUOTE: CRFs can also be viewed simply as Markov networks with exponential form potentials, as in section(9.6.4).
2009
- http://www.di.ens.fr/~mschmidt/Software/UGM.html
- UGM is a set of Matlab functions implementing various tasks in probabilistic undirected graphical models of discrete data with pairwise (and unary) potentials. Specifically, it implements a variety of methods for the following four tasks:
- Decoding: Computing the most likely configuration.
- Inference: Computing the partition function and marginal probabilities.
- Sampling: Generating samples from the distribution.
- Parameter Estimation: Given data, computing maximum likelihood (or MAP) estimates of the parameters.
- UGM is a set of Matlab functions implementing various tasks in probabilistic undirected graphical models of discrete data with pairwise (and unary) potentials. Specifically, it implements a variety of methods for the following four tasks:
2008
- Doina Precup. (2008). "Lecture 5: Undirected graphical models."
- QUOTE: An undirected graph over a set of random variables <math>{X_1,...,X_n}<math> is called a undirected graphical model or Markov random field (MRF) or Markov network ...
* Undirected models are neither more nor less expressive than directed models; they are simply different.
* The semantics of an undirected model naturally capture correlation of r.v.s, not causation
* If you ever want, in an application, to write a Bayes net with, cycles, it is a sign that the right model is undirected.
- QUOTE: An undirected graph over a set of random variables <math>{X_1,...,X_n}<math> is called a undirected graphical model or Markov random field (MRF) or Markov network ...