A Variational Inference Procedure Allowing Internal Structure for Overlapping Clusters and Deterministic Constraints

Dan Geiger, Christopher Meek, Ydo Wexler

JAIR 2006 pp. 1-23

doi:10.1613/JAIR.2028 /jair/2006/geiger2006jair-variational/

Abstract

We develop a novel algorithm, called VIP*, for structured variational approximate inference. This algorithm extends known algorithms to allow efficient multiple potential updates for overlapping clusters, and overcomes the difficulties imposed by deterministic constraints. The algorithm's convergence is proven and its applicability demonstrated for genetic linkage analysis.

PDF JAIR Semantic Scholar

Cite

Text

Geiger et al. "A Variational Inference Procedure Allowing Internal Structure for Overlapping Clusters and Deterministic Constraints." Journal of Artificial Intelligence Research, 2006. doi:10.1613/JAIR.2028

Markdown

[Geiger et al. "A Variational Inference Procedure Allowing Internal Structure for Overlapping Clusters and Deterministic Constraints." Journal of Artificial Intelligence Research, 2006.](https://mlanthology.org/jair/2006/geiger2006jair-variational/) doi:10.1613/JAIR.2028

BibTeX

@article{geiger2006jair-variational,
  title     = {{A Variational Inference Procedure Allowing Internal Structure for Overlapping Clusters and Deterministic Constraints}},
  author    = {Geiger, Dan and Meek, Christopher and Wexler, Ydo},
  journal   = {Journal of Artificial Intelligence Research},
  year      = {2006},
  pages     = {1-23},
  doi       = {10.1613/JAIR.2028},
  volume    = {27},
  url       = {https://mlanthology.org/jair/2006/geiger2006jair-variational/}
}