Learning mixed graphical models from data with p larger than n
This work addresses a gap in structure learning for mixed graphical models in high-dimensional data, which is relevant for fields like bioinformatics or social sciences where such data is common, but it appears incremental as it extends existing approaches to a new setting.
The authors tackled the problem of learning the structure of mixed graphical models (with both discrete and continuous variables) in high-dimensional settings where the number of variables p exceeds the sample size n, a scenario previously underexplored. They proposed a statistical learning procedure based on limited-order correlations and evaluated it on synthetic and real data, though no concrete performance numbers were provided in the abstract.
Structure learning of Gaussian graphical models is an extensively studied problem in the classical multivariate setting where the sample size n is larger than the number of random variables p, as well as in the more challenging setting when p>>n. However, analogous approaches for learning the structure of graphical models with mixed discrete and continuous variables when p>>n remain largely unexplored. Here we describe a statistical learning procedure for this problem based on limited-order correlations and assess its performance with synthetic and real data.