Methods for learning sparse Bayesian networks and other graphical models from high-dimensional data via sparse regularization. Designed to handle:
- Experimental data with interventions
- Mixed observational / experimental data
- High-dimensional data with p >> n
- Datasets with thousands of variables (tested up to p=8000)
- Continuous and discrete data
The workhorse behind sparsebn is the sparsebnUtils package, which provides various S3 classes and methods for representing and manipulating graphs. The basic algorithms are implemented in ccdrAlgorithm and discretecdAlgorithm.
The main methods for learning graphical models are:
estimate.dagfor directed acyclic graphs (Bayesian networks).estimate.precisionfor undirected graphs (Markov random fields).estimate.covariancefor covariance matrices.
Currently, estimation of precision and covariances matrices is limited to Gaussian data.
You can install:
-
the latest CRAN version with
install.packages("sparsebn") -
the latest development version from GitHub with
devtools::install_github(c("itsrainingdata/sparsebn/", "itsrainingdata/sparsebnUtils/dev", "itsrainingdata/ccdrAlgorithm/dev", "gujyjean/discretecdAlgorithm"))
[1] Aragam, B. and Zhou, Q. (2015). Concave penalized estimation of sparse Gaussian Bayesian networks. The Journal of Machine Learning Research. 16(Nov):2273−2328.
[2] Fu, F., Gu, J., and Zhou, Q. (2014). Adaptive penalized estimation of directed acyclic graphs from categorical data. arXiv: 1403.2310.
[3] Aragam, B., Amini, A. A., and Zhou, Q. (2015). Learning directed acyclic graphs with penalized neighbourhood regression. arXiv: 1511.08963.
[4] Fu, F. and Zhou, Q. (2013). Learning sparse causal Gaussian networks with experimental intervention: Regularization and coordinate descent. Journal of the American Statistical Association, 108: 288-300.