Learning Hierarchical Relational Representations through Relational Convolutions

TL;DR

This paper introduces relational convolutional networks that learn hierarchical relational features by convolving graphlet filters, enabling the modeling of complex, higher-order relations in data.

Contribution

The paper proposes a novel neural architecture with relational convolutions and graphlet filters to effectively learn hierarchical relational representations.

Findings

Relational convolutional networks outperform baseline models on hierarchical relational tasks.

The architecture captures complex relational patterns through composition of simple modules.

Experimental results demonstrate the effectiveness of the proposed method in modeling hierarchical relations.

Abstract

An evolving area of research in deep learning is the study of architectures and inductive biases that support the learning of relational feature representations. In this paper, we address the challenge of learning representations of hierarchical relations--that is, higher-order relational patterns among groups of objects. We introduce "relational convolutional networks", a neural architecture equipped with computational mechanisms that capture progressively more complex relational features through the composition of simple modules. A key component of this framework is a novel operation that captures relational patterns in groups of objects by convolving graphlet filters--learnable templates of relational patterns--against subsets of the input. Composing relational convolutions gives rise to a deep architecture that learns representations of higher-order, hierarchical relations. We present the motivation and details of the architecture, together with a set of experiments to demonstrate how relational convolutional networks can provide an effective framework for modeling relational tasks that have hierarchical structure.