Beyond Graph Neural Networks with Lifted Relational Neural Networks

TL;DR

This paper introduces a declarative differentiable programming framework using Lifted Relational Neural Networks, enabling flexible, compact encoding of graph neural networks and extending their relational expressiveness.

Contribution

It presents a novel declarative approach for relational learning with neural networks, allowing dynamic unfolding of differentiable graphs and easier extension of GNN models.

Findings

Efficient encoding of diverse neural architectures using the framework

Comparison shows competitive correctness and efficiency against specialized GNN frameworks

Framework facilitates higher relational expressiveness in GNN models

Abstract

We demonstrate a declarative differentiable programming framework based on the language of Lifted Relational Neural Networks, where small parameterized logic programs are used to encode relational learning scenarios. When presented with relational data, such as various forms of graphs, the program interpreter dynamically unfolds differentiable computational graphs to be used for the program parameter optimization by standard means. Following from the used declarative Datalog abstraction, this results into compact and elegant learning programs, in contrast with the existing procedural approaches operating directly on the computational graph level. We illustrate how this idea can be used for an efficient encoding of a diverse range of existing advanced neural architectures, with a particular focus on Graph Neural Networks (GNNs). Additionally, we show how the contemporary GNN models can be easily extended towards higher relational expressiveness. In the experiments, we demonstrate correctness and computation efficiency through comparison against specialized GNN deep learning frameworks, while shedding some light on the learning performance of existing GNN models.