Abstract:Graph neural networks (GNNs) have gained significant attention in recent years for their ability to process data that may be represented as graphs. This has prompted several studies to explore their representational capability based on the graph isomorphism task. These works inherently assume a countable node feature representation, potentially limiting their applicability. Interestingly, only a few study GNNs with uncountable node feature representation. In the paper, a novel perspective on the representational capability of GNNs is investigated across all levels$\unicode{x2014}$node-level, neighborhood-level, and graph-level$\unicode{x2014}$when the space of node feature representation is uncountable. More specifically, the strict injective and metric requirements are softly relaxed by employing a pseudometric distance on the space of input to create a soft-injective function such that distinct inputs may produce similar outputs if and only if the pseudometric deems the inputs to be sufficiently similar on some representation. As a consequence, a simple and computationally efficient soft-isomorphic relational graph convolution network (SIR-GCN) that emphasizes the contextualized transformation of neighborhood feature representations via anisotropic and dynamic message functions is proposed. A mathematical discussion on the relationship between SIR-GCN and widely used GNNs is then laid out to put the contribution into context, establishing SIR-GCN as a generalization of classical GNN methodologies. Experiments on synthetic and benchmark datasets then demonstrate the relative superiority of SIR-GCN, outperforming comparable models in node and graph property prediction tasks.

Discovering the Representation Bottleneck of Graph Neural Networks

Discovering the Representation Bottleneck of Graph Neural Networks from Multi-order Interactions

Discovering and Explaining the Representation Bottleneck of Graph Neural Networks from Multi-order Interactions

Locality-Aware Graph-Rewiring in GNNs

Dynamic Graph Information Bottleneck

Breaking the Bottleneck on Graphs with Structured State Spaces

How Powerful are Graph Neural Networks?

Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting

Commute-Time-Optimised Graphs for GNNs

A network analysis-based framework to understand the representation dynamics of graph neural networks

What Do GNNs Actually Learn? Towards Understanding their Representations

Rethinking and Extending the Probabilistic Inference Capacity of GNNs

Rewiring Techniques to Mitigate Oversquashing and Oversmoothing in GNNs: A Survey

Revisiting graph neural networks from hybrid regularized graph signal reconstruction

Bipartite Graph Neural Networks for Efficient Node Representation Learning

Contextualized Messages Boost Graph Representations

Graph Neural Networks Inspired by Classical Iterative Algorithms

InterpGNN: Understand and Improve Generalization Ability of Transdutive GNNs Through the Lens of Interplay Between Train and Test Nodes

From Node Interaction to Hop Interaction: New Effective and Scalable Graph Learning Paradigm

Graph Representation Learning with Individualization and Refinement

Incorporating Retrieval-based Causal Learning with Information Bottlenecks for Interpretable Graph Neural Networks