Efficient Subgraph GNNs by Learning Effective Selection Policies
This addresses a scalability bottleneck for graph neural networks, enabling more efficient and expressive models, though it is incremental as it builds on existing subgraph GNN frameworks.
The paper tackles the computational inefficiency of Subgraph GNNs by learning data-driven subgraph selection policies, proving that small subsets can distinguish graphs and showing that Policy-Learn outperforms baselines across multiple datasets.
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.