Topology-Informed Graph Transformer
This work addresses a key bottleneck in graph transformers for researchers and practitioners in graph learning, though it appears incremental as it builds on existing transformer architectures.
The paper tackled the challenge of enhancing graph transformers' ability to distinguish graph isomorphisms, which is crucial for predictive performance, by introducing the Topology-Informed Graph Transformer (TIGT). TIGT outperformed previous graph transformers on a synthetic dataset for isomorphism classification and showed competitive results on various benchmarks.
Transformers have revolutionized performance in Natural Language Processing and Vision, paving the way for their integration with Graph Neural Networks (GNNs). One key challenge in enhancing graph transformers is strengthening the discriminative power of distinguishing isomorphisms of graphs, which plays a crucial role in boosting their predictive performances. To address this challenge, we introduce 'Topology-Informed Graph Transformer (TIGT)', a novel transformer enhancing both discriminative power in detecting graph isomorphisms and the overall performance of Graph Transformers. TIGT consists of four components: A topological positional embedding layer using non-isomorphic universal covers based on cyclic subgraphs of graphs to ensure unique graph representation: A dual-path message-passing layer to explicitly encode topological characteristics throughout the encoder layers: A global attention mechanism: And a graph information layer to recalibrate channel-wise graph features for better feature representation. TIGT outperforms previous Graph Transformers in classifying synthetic dataset aimed at distinguishing isomorphism classes of graphs. Additionally, mathematical analysis and empirical evaluations highlight our model's competitive edge over state-of-the-art Graph Transformers across various benchmark datasets.