Bayesian Out-Trees
This provides a method for handling complex dependencies in data, such as in taxonomy and phylogenetic datasets, though it appears incremental as it builds on existing graph-based Bayesian frameworks.
The paper tackles the problem of modeling non-iid data with latent directed graph structures by proposing a Bayesian approach using out-tree graphs, resulting in efficient inference algorithms for unsupervised and semi-supervised learning on datasets like UCI and taxonomies.
A Bayesian treatment of latent directed graph structure for non-iid data is provided where each child datum is sampled with a directed conditional dependence on a single unknown parent datum. The latent graph structure is assumed to lie in the family of directed out-tree graphs which leads to efficient Bayesian inference. The latent likelihood of the data and its gradients are computable in closed form via Tutte's directed matrix tree theorem using determinants and inverses of the out-Laplacian. This novel likelihood subsumes iid likelihood, is exchangeable and yields efficient unsupervised and semi-supervised learning algorithms. In addition to handling taxonomy and phylogenetic datasets the out-tree assumption performs surprisingly well as a semi-parametric density estimator on standard iid datasets. Experiments with unsupervised and semisupervised learning are shown on various UCI and taxonomy datasets.