paper-with-me

홈 › Papers

Unifying approach to uniform expressivity of graph neural networks

2026-02-20 · Huan Luo, Jonni Virtema arxiv

The expressive power of Graph Neural Networks (GNNs) is often analysed via correspondence to the Weisfeiler-Leman (WL) algorithm and fragments of first-order logic. Standard GNNs are limited to performing aggregation over immediate neighbourhoods or over global read-outs. To increase their expressivity, recent attempts have been made to incorporate substructural information (e.g. cycle counts and subgraph properties). In this paper, we formalize this architectural trend by introducing Template GNNs (T-GNNs), a generalized framework where node features are updated by aggregating over valid template embeddings from a specified set of graph templates. We propose a corresponding logic, Graded template modal logic (GML(T)), and generalized notions of template-based bisimulation and WL algorithm. We establish an equivalence between the expressive power of T-GNNs and GML(T), and provide a unifying approach for analysing GNN expressivity: we show how standard AC-GNNs and its recent variants can be interpreted as instantiations of T-GNNs.

📄 PDF Abstract BibTeX arXiv:2602.18409

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Is uniform expressivity too restrictive? Towards efficient expressivity of graph neural networks

2024-10-02 · Sammy Khalife, Josué Tonelli-Cueto

Uniform expressivity guarantees that a Graph Neural Network (GNN) can express a query without the parameters depending on the size of the input graphs. This property is desirable in applications in order to have number o…

Graph Neural Network

A Unifying Relational Perspective on Expressive Lottery Tickets

2026-08-19 · Lorenz Kummer, Samir Moustafa, Anatol Ehrlich, Franka Bause 외 arxiv

Graph neural networks (GNNs) are widely used, but how parameter sparsity affects the expressivity of relational (RGNNs) and temporal (TGNNs) variants is poorly understood. The Strong Expressive Lottery Ticket Hypothesis …

Distinguished In Uniform: Self Attention Vs. Virtual Nodes

2024-05-20 · Eran Rosenbluth, Jan Tönshoff, Martin Ritzert, Berke Kisin 외

Graph Transformers (GTs) such as SAN and GPS are graph processing models that combine Message-Passing GNNs (MPGNNs) with global Self-Attention. They were shown to be universal function approximators, with two reservation…

Are Targeted Messages More Effective?

2024-03-11 · Martin Grohe, Eran Rosenbluth

Graph neural networks (GNN) are deep learning architectures for graphs. Essentially, a GNN is a distributed message passing algorithm, which is controlled by parameters learned from data. It operates on the vertices of a…

Invariant-Stratified Propagation for Expressive Graph Neural Networks

2026-03-02 · Asela Hevapathige, Ahad N. Zehmakan, Asiri Wijesinghe, Saman Halgamuge arxiv

Graph Neural Networks (GNNs) face fundamental limitations in expressivity and capturing structural heterogeneity. Standard message-passing architectures are constrained by the 1-dimensional Weisfeiler-Leman (1-WL) test, …

Graph ClassificationNode Classification