Content deleted Content added
m →The Chinese restaurant process: out-of-place remark |
m The rows contain the repetitions, not the columns. |
||
Line 1:
[[File:Dirichlet process draws.svg|thumb|300px|Draws from the Dirichlet process <math>\mathrm{DP}\left(N(0,1), \alpha\right)</math>. The four rows use different <math>\alpha</math> (top to bottom: 1, 10, 100 and 1000) and each
In [[probability theory]], '''Dirichlet processes''' (after [[Peter Gustav Lejeune Dirichlet]]) are a family of [[stochastic process]]es whose [[realization (probability)|realizations]] are [[probability distribution]]s. In other words, a Dirichlet process is a probability distribution whose range is itself a set of probability distributions. It is often used in [[Bayesian inference]] to describe the [[prior probability|prior]] knowledge about the distribution of [[random variable]]s—how likely it is that the random variables are distributed according to one or another particular distribution.
| |||