1

We propose a probabilistic model that bridges graph machine learning and sum-product networks to tractably answer probabilistic queries. Our Graph-induced Sum-Product Network can solve unsupervised and supervised graph tasks and it is especially effective in modeling the missing data distribution as well as exploiting unlabeled data in a scarce supervision scenario.

In this paper I try to address the following question: “Is it really worth to build a k-Nearest Neighbor graph from tabular data and then apply deep graph nets on top of it to classify each sample in the table?

Who said HMMs cannot be adapted for graph-structured data? In this work, we extend CGMM to the temporal domain, and we report performances close to other popular neural models with a probabilistic graph embedding model!

We investigate the quality of quantitative measures that assess the utility of a graph structure for node classification tasks.

Combining Deep Graph Networks with Bayesian nonparametric techniques to unsupervised learning of node/graph representations while automatizing the choice of the hyper-parameters during training.

Combining Deep Graph Networks with Mixture Density Networks to model multimodal output distributions conditioned on arbitrary input graphs.