A Spatially-Constrained Normalized Gamma Process for Data Clustering - Archive ouverte HAL Access content directly
Conference Papers Year : 2012

A Spatially-Constrained Normalized Gamma Process for Data Clustering

(1) , (2) , (2)
1
2

Abstract

In this work, we propose a novel nonparametric Bayesian method for clustering of data with spatial interdependencies. Specifically, we devise a novel normalized Gamma process, regulated by a simplified (pointwise) Markov random field (Gibbsian) distribution with a countably infinite number of states. As a result of its construction, the proposed model allows for introducing spatial dependencies in the clustering mechanics of the normalized Gamma process, thus yielding a novel nonparametric Bayesian method for spatial data clustering. We derive an efficient truncated variational Bayesian algorithm for model inference. We examine the efficacy of our approach by considering an image segmentation application using a real-world dataset. We show that our approach outperforms related methods from the field of Bayesian nonparametrics, including the infinite hidden Markov random field model, and the Dirichlet process prior.
Fichier principal
Vignette du fichier
978-3-642-33409-2_35_Chapter.pdf (660.88 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01521423 , version 1 (11-05-2017)

Licence

Attribution - CC BY 4.0

Identifiers

Cite

Sotirios P. Chatzis, Dimitrios Korkinof, Yiannis Demiris. A Spatially-Constrained Normalized Gamma Process for Data Clustering. 8th International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2012, Halkidiki, Greece. pp.337-346, ⟨10.1007/978-3-642-33409-2_35⟩. ⟨hal-01521423⟩
59 View
89 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More