Deep Models for Relational Databases

Due to its extensive use in databases, the relational model is ubiquitous in representing big-data. We apply deep learning to relational data(bases) by introducing an Equivariant Relational Layer (ERL), a neural network layer derived from the entity-relationship model. Our layer relies on the identification of exchangeabilities in the relational data, and their expression as a permutation group. We prove that an ERL is an optimal feed-forward layer under the given exchangeability constraints, and subsumes recently introduced equivariant deep models for sets, exchangeable tensors, and graphs. The proposed feed-forward layer has linear complexity in the size of the data and can be used for both inductive and transductive reasoning about databases, including database embedding, and the prediction of missing records. This opens the door to the application of deep learning to one of the most abundant forms of data.