Variational message passing (VMP) is an approximate inference technique for continuous- or discrete-valued Bayesian networks, with conjugate-exponential parents, developed by John Winn. VMP was developed as a means of generalizing the approximate variational methods used by such techniques as Latent Dirichlet allocation and works by updating an approximate distribution at each node through messages in the node's Markov blanket.
Read more about Variational Message Passing: Likelihood Lower Bound, Determining The Update Rule, Messages in Variational Message Passing, Relationship To Exponential Families, VMP Algorithm, Constraints
Famous quotes containing the words message and/or passing:
“What the hell is nostalgia doing in a science-fiction film? With the whole universe and all the future to play in, Lucas took his marvelous toys and crawled under the fringed cloth on the parlor table, back into a nice safe hideyhole, along with Flash Gordon and the Cowardly Lion and Luck Skywalker and the Flying Aces and the Hitler Jugend. If there’s a message there, I don’t think I want to hear it.”
—Ursula K. Le Guin (b. 1929)
“There is, indeed, nothing that so much seduces reason from vigilance, as the thought of passing life with an amiable woman.”
—Samuel Johnson (1709–1784)