Correlated Non-Parametric Latent Feature Models

Abstract

We are often interested in explaining data through a set of hidden factors or features. When the number of hidden features is unknown, the Indian Buffet Process (IBP) is a nonparametric latent feature model that does not bound the number of active features in dataset. However, the IBP assumes that all latent features are uncorrelated, making it inadequate for many realworld problems. We introduce a framework for correlated nonparametric feature models, generalising the IBP. We use this framework to generate several specific models and demonstrate applications on realworld datasets.

Appears in Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence (UAI2009)

References in corpus (1)

• Nonparametric Bayes Pachinko Allocation

Cited by in corpus (10)

• Role Discovery in Networks
• Nonparametric Bayesian sparse factor models with application to gene expression modeling
• Economic Complexity Unfolded: Interpretable Model for the Productive Structure of Economies
• Central limit theorems for an Indian buffet model with random weights
• Correlated Random Measures
• Distance Dependent Infinite Latent Feature Models
• Beta diffusion trees and hierarchical feature allocations
• Weakly Supervised Image Annotation and Segmentation with Objects and Attributes
• Restricted Indian Buffet Processes
• A latent-observed dissimilarity measure