Learning directed acyclic graph SPNs in sub-quadratic time

Abstract

In this paper, we present Prometheus, a graph partitioning based algorithm that creates multiple variable decompositions efficiently for learning Sum-Product Network structures across both continuous and discrete domains. Prometheus proceeds by creating multiple candidate decompositions that are represented compactly with an acyclic directed graph in which common parts of different decompositions are shared. It eliminates the correlation threshold hyperparameter often used in other structure learning techniques, allowing Prometheus to learn structures that are robust in low data regimes. Prometheus outperforms other structure learning techniques in 30 discrete and continuous domains. We also extend Prometheus to exploit sparsity in correlations between features in order to obtain an efficient sub-quadratic algorithm (w.r.t. the number of features) that scales better to high dimensional datasets.