Bayesian nonparametrics

E1031259

branch of Bayesian statistics statistical methodology subfield of nonparametric statistics

Bayesian nonparametrics is a branch of Bayesian statistics that uses flexible, potentially infinite-dimensional models to let data determine model complexity rather than fixing a finite set of parameters in advance.

Try in SPARQL Jump to: Statements Referenced by

Statements (56)

Predicate	Object
instanceOf	branch of Bayesian statistics ⓘ statistical methodology ⓘ subfield of nonparametric statistics ⓘ
contrastsWith	frequentist nonparametric methods ⓘ parametric Bayesian statistics ⓘ
fieldOfStudy	machine learning ⓘ statistics ⓘ
hasAdvantage	automatically adapts model complexity ⓘ can capture complex data structures ⓘ can model an unbounded number of clusters ⓘ provides full Bayesian uncertainty quantification ⓘ
hasApplication	clustering ⓘ density estimation ⓘ graphical models ⓘ hierarchical modeling ⓘ latent feature modeling ⓘ mixture modeling ⓘ nonlinear function estimation ⓘ regression ⓘ survival analysis ⓘ time series modeling ⓘ topic modeling ⓘ
hasCharacteristic	allows model complexity to grow with data ⓘ avoids fixing the number of parameters in advance ⓘ supports flexible clustering structures ⓘ supports flexible density estimation ⓘ supports flexible function estimation ⓘ uses infinite-dimensional parameter spaces ⓘ uses stochastic processes as priors ⓘ
hasGoal	let data determine model complexity ⓘ
hasMethod	Chinese restaurant franchise ⓘ Chinese restaurant process ⓘ Dirichlet process NERFINISHED ⓘ Dirichlet process mixture model NERFINISHED ⓘ Dirichlet process mixture of Gaussians NERFINISHED ⓘ Gaussian process NERFINISHED ⓘ Gaussian process regression ⓘ Indian buffet process NERFINISHED ⓘ Indian buffet process latent feature model ⓘ Pitman–Yor process NERFINISHED ⓘ beta process ⓘ hierarchical Dirichlet process NERFINISHED ⓘ normalized random measures ⓘ
relatedTo	Bayesian machine learning NERFINISHED ⓘ nonparametric Bayes ⓘ probabilistic modeling ⓘ
usesConcept	Bayesian inference ⓘ Chinese restaurant process ⓘ Gibbs sampling NERFINISHED ⓘ Markov chain Monte Carlo NERFINISHED ⓘ exchangeability ⓘ posterior distribution ⓘ prior distribution ⓘ stick-breaking construction ⓘ stochastic process priors ⓘ variational inference ⓘ

Referenced by (2)

Full triples — surface form annotated when it differs from this entity's canonical label.

Stick-breaking construction for the Indian buffet process → mainTopic → Bayesian nonparametrics ⓘ

Kolmogorov extension theorem → usedIn → Bayesian nonparametrics ⓘ