the Indian Buffet Process: An Introduction and Review: Difference between revisions

From statwiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 5: Line 5:
==Introduction==
==Introduction==
IBP is often used in factor analysis as a prior of infinite factors.
IBP is often used in factor analysis as a prior of infinite factors.
IBP can be viewed as an extension of DP, where we drop the constraint <math> \sum_(i=1)^(\inf){\pi_i}=1 </math> .   
IBP can be viewed as an extension of DP, where we drop the constraint <math> \sum_{i=1}^{\inf}{\pi_i}=1 </math> .   


==Representations==
==Representations==

Revision as of 20:15, 9 August 2013

The Indian Buffet Process (IBP) is one of Bayesian nonparametric models, which is a prior measure on an infinite binary matrix. Unlike the Dirichlet process(DP), where each atom has negative correlation, IBP assumes each atom is independent.


Introduction

IBP is often used in factor analysis as a prior of infinite factors. IBP can be viewed as an extension of DP, where we drop the constraint [math]\displaystyle{ \sum_{i=1}^{\inf}{\pi_i}=1 }[/math] .

Representations

the limiting of finite distribution on sparse binary feature matrices

the Indian buffet metaphor

comparison

Inference

Conclusion