Constraints
Because every child must be conjugate to its parent, this limits the types of distributions that can be used in the model. For example, the parents of a Gaussian distribution must be a Gaussian distribution (corresponding to the Mean) and a gamma distribution (corresponding to the precision, or one over in more common paramaterizations). Discrete variables can have Dirichlet parents, and Poisson and exponential nodes must have gamma parents. However, if the data can be modeled in this manner, VMP offers a generalized framework for providing inference.
Read more about this topic: Variational Message Passing
Famous quotes containing the word constraints:
“The analogy between the mind and a computer fails for many reasons. The brain is constructed by principles that assure diversity and degeneracy. Unlike a computer, it has no replicative memory. It is historical and value driven. It forms categories by internal criteria and by constraints acting at many scales, not by means of a syntactically constructed program. The world with which the brain interacts is not unequivocally made up of classical categories.”
—Gerald M. Edelman (b. 1928)
“Play is a major avenue for learning to manage anxiety. It gives the child a safe space where she can experiment at will, suspending the rules and constraints of physical and social reality. In play, the child becomes master rather than subject.... Play allows the child to transcend passivity and to become the active doer of what happens around her.”
—Alicia F. Lieberman (20th century)