Abstract:Graph neural networks (GNNs) have achieved remarkable success in a variety of machine learning tasks over graph data. Existing GNNs usually rely on message passing, i.e., computing node representations by gathering information from the neighborhood, to build their underlying computational graphs. They are known fairly limited in expressive power, and often fail to capture global characteristics of graphs. To overcome the issue, a popular solution is to use Laplacian eigenvectors as additional node features, as they contain global positional information of nodes, and can serve as extra node identifiers aiding GNNs to separate structurally similar nodes. For such an approach, properly handling the orthogonal group symmetry among eigenvectors with equal eigenvalue is crucial for its stability and generalizability. However, using a naive orthogonal group invariant encoder for each separate eigenspace may not keep the full expressivity in the Laplacian eigenvectors. Moreover, computing such invariants inevitably entails a hard split of Laplacian eigenvalues according to their numerical identity, which suffers from great instability when the graph structure is perturbed. In this paper, we propose a novel method exploiting Laplacian eigenvectors to generate stable and globally expressive graph representations. The main difference from previous works is that (i) our method utilizes learnable orthogonal group invariant representations for each Laplacian eigenspace, based upon powerful orthogonal group equivariant neural network layers already well studied in the literature, and that (ii) our method deals with numerically close eigenvalues in a smooth fashion, ensuring its better robustness against perturbations. Experiments on various graph learning benchmarks witness the competitive performance of our method, especially its great potential to learn global properties of graphs.

Graph Neural Network Meets Sparse Representation: Graph Sparse Neural Networks Via Exclusive Group Lasso

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

Robust Graph Representation Learning Via Neural Sparsification.

Beyond smoothness: A general optimization framework for graph neural networks with negative Laplacian regularization

Higher-Order GNNs Meet Efficiency: Sparse Sobolev Graph Neural Networks

Enhanced Expressivity in Graph Neural Networks with Lanczos-Based Linear Constraints

Discovering the Representation Bottleneck of Graph Neural Networks

Graph Representation Learning on Noise and Sparse Labels.

Two Heads Are Better Than One: Boosting Graph Sparse Training via Semantic and Topological Awareness

Graph Sparsification via Mixture of Graphs

RSGNN: residual structure graph neural network

Deep Graph Neural Networks via Flexible Subgraph Aggregation

Learning the Network of Graphs for Graph Neural Networks

GraphSAGE++: Weighted Multi-scale GNN for Graph Representation Learning

GraLSP: Graph Neural Networks with Local Structural Patterns

Signed Graph Neural Network with Latent Groups.

Contrastive Message Passing for Robust Graph Neural Networks with Sparse Labels

Towards Robust Graph Neural Networks for Noisy Graphs with Sparse Labels

Towards Stable, Globally Expressive Graph Representations with Laplacian Eigenvectors

Group And Graph Joint Sparsity For Linked Data Classification

Graph Neural Networks Inspired by Classical Iterative Algorithms