Efficient Subgraph GNNs by Learning Effective Selection Policies

TL;DR

This paper introduces Policy-Learn, a method for learning effective subgraph selection policies in Subgraph GNNs, significantly reducing computational costs while maintaining high expressiveness and outperforming existing methods.

Contribution

We propose a novel learning-based approach, Policy-Learn, that efficiently selects subgraphs for GNNs, enabling scalable and expressive graph representations.

Findings

Policy-Learn outperforms baselines on multiple datasets.

It can identify small subgraph subsets that distinguish WL-indistinguishable graphs.

The approach learns efficient selection policies unlike random or prior methods.

Abstract

Subgraph GNNs are provably expressive neural architectures that learn graph representations from sets of subgraphs. Unfortunately, their applicability is hampered by the computational complexity associated with performing message passing on many subgraphs. In this paper, we consider the problem of learning to select a small subset of the large set of possible subgraphs in a data-driven fashion. We first motivate the problem by proving that there are families of WL-indistinguishable graphs for which there exist efficient subgraph selection policies: small subsets of subgraphs that can already identify all the graphs within the family. We then propose a new approach, called Policy-Learn, that learns how to select subgraphs in an iterative manner. We prove that, unlike popular random policies and prior work addressing the same problem, our architecture is able to learn the efficient policies mentioned above. Our experimental results demonstrate that Policy-Learn outperforms existing baselines across a wide range of datasets.