Automatic Relation-aware Graph Network Proliferation

TL;DR

This paper introduces ARGNP, a novel method for automatically designing relation-aware GNN architectures by exploring a dual relation-aware search space and employing a network proliferation strategy, leading to superior performance.

Contribution

The paper proposes a new dual relation-aware search space and a network proliferation approach for efficient GNN architecture search, addressing limitations of existing methods.

Findings

GNNs found by ARGNP outperform state-of-the-art models on six datasets.

The relation-guided message passing mechanism improves hierarchical relational learning.

The proliferation search paradigm effectively explores complex GNN architectures.

Abstract

Graph neural architecture search has sparked much attention as Graph Neural Networks (GNNs) have shown powerful reasoning capability in many relational tasks. However, the currently used graph search space overemphasizes learning node features and neglects mining hierarchical relational information. Moreover, due to diverse mechanisms in the message passing, the graph search space is much larger than that of CNNs. This hinders the straightforward application of classical search strategies for exploring complicated graph search space. We propose Automatic Relation-aware Graph Network Proliferation (ARGNP) for efficiently searching GNNs with a relation-guided message passing mechanism. Specifically, we first devise a novel dual relation-aware graph search space that comprises both node and relation learning operations. These operations can extract hierarchical node/relational information and provide anisotropic guidance for message passing on a graph. Second, analogous to cell proliferation, we design a network proliferation search paradigm to progressively determine the GNN architectures by iteratively performing network division and differentiation. The experiments on six datasets for four graph learning tasks demonstrate that GNNs produced by our method are superior to the current state-of-the-art hand-crafted and search-based GNNs. Codes are available at https://github.com/phython96/ARGNP.