Learning Measurement Models for Unobserved Variables
Abstract
Observed associations in a database may be due in whole or part to variations in unrecorded ("latent") variables. Identifying such variables and their causal relationships with one another is a principal goal in many scientific and practical domains. Previous work shows that, given a partition of observed variables such that members of a class share only a single latent common cause, standard search algorithms for causal Bayes nets can infer structural relations between latent variables. We introduce an algorithm for discovering such partitions when they exist. Uniquely among available procedures, the algorithm is (asymptotically) correct under standard assumptions in causal Bayes net search algorithms, requires no prior knowledge of the number of latent variables, and does not depend on the mathematical form of the relationships among the latent variables. We evaluate the algorithm on a variety of simulated data sets.
Cite
Text
de Andrade e Silva et al. "Learning Measurement Models for Unobserved Variables." Conference on Uncertainty in Artificial Intelligence, 2003.Markdown
[de Andrade e Silva et al. "Learning Measurement Models for Unobserved Variables." Conference on Uncertainty in Artificial Intelligence, 2003.](https://mlanthology.org/uai/2003/deandradeesilva2003uai-learning/)BibTeX
@inproceedings{deandradeesilva2003uai-learning,
title = {{Learning Measurement Models for Unobserved Variables}},
author = {de Andrade e Silva, Ricardo Bezerra and Scheines, Richard and Glymour, Clark and Spirtes, Peter},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2003},
pages = {543-550},
url = {https://mlanthology.org/uai/2003/deandradeesilva2003uai-learning/}
}