On the Relationship Between Robustness and Expressivity of Graph Neural Networks

• URL: http://arxiv.org/abs/2504.13786v1
• Date: Fri, 18 Apr 2025 16:38:33 GMT
• Title: On the Relationship Between Robustness and Expressivity of Graph Neural Networks
• Authors: Lorenz Kummer, Wilfried N. Gansterer, Nils M. Kriege,
• Abstract summary: Graph Neural Networks (GNNs) are vulnerable to bit-flip attacks (BFAs)<n>We introduce an analytical framework to study the influence of architectural features, graph properties, and their interaction.<n>We derive theoretical bounds for the number of bit flips required to degrade GNN expressivity on a dataset.
• Score: 7.161966906570077
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We investigate the vulnerability of Graph Neural Networks (GNNs) to bit-flip attacks (BFAs) by introducing an analytical framework to study the influence of architectural features, graph properties, and their interaction. The expressivity of GNNs refers to their ability to distinguish non-isomorphic graphs and depends on the encoding of node neighborhoods. We examine the vulnerability of neural multiset functions commonly used for this purpose and establish formal criteria to characterize a GNN's susceptibility to losing expressivity due to BFAs. This enables an analysis of the impact of homophily, graph structural variety, feature encoding, and activation functions on GNN robustness. We derive theoretical bounds for the number of bit flips required to degrade GNN expressivity on a dataset, identifying ReLU-activated GNNs operating on highly homophilous graphs with low-dimensional or one-hot encoded features as particularly susceptible. Empirical results using ten real-world datasets confirm the statistical significance of our key theoretical insights and offer actionable results to mitigate BFA risks in expressivity-critical applications.

Related papers

• On the Computational Capability of Graph Neural Networks: A Circuit Complexity Bound Perspective [28.497567290882355]
  Graph Neural Networks (GNNs) have become the standard approach for learning and reasoning over relational data. This paper explores the computational limitations of GNNs through the lens of circuit complexity. Specifically, we analyze the circuit complexity of common GNN architectures and prove that under constraints of constant-depth layers, linear or sublinear embedding sizes, and precision, GNNs cannot solve key problems such as graph connectivity and graph isomorphism.
  arXiv  Detail & Related papers  (2025-01-11T05:54:10Z)
• On the Impact of Feature Heterophily on Link Prediction with Graph Neural Networks [12.26334940017605]
  Heterophily, or the tendency of connected nodes in networks to have different class labels or dissimilar features, has been identified as challenging for many Graph Neural Network (GNN) models. We focus on the link prediction task and systematically analyze the impact of heterophily in node features on GNN performance.
  arXiv  Detail & Related papers  (2024-09-26T02:19:48Z)
• On the Topology Awareness and Generalization Performance of Graph Neural Networks [6.598758004828656]
  We introduce a comprehensive framework to characterize the topology awareness of GNNs across any topological feature. We conduct a case study using the intrinsic graph metric the shortest path distance on various benchmark datasets.
  arXiv  Detail & Related papers  (2024-03-07T13:33:30Z)
• Rethinking Causal Relationships Learning in Graph Neural Networks [24.7962807148905]
  We introduce a lightweight and adaptable GNN module designed to strengthen GNNs' causal learning capabilities. We empirically validate the effectiveness of the proposed module.
  arXiv  Detail & Related papers  (2023-12-15T08:54:32Z)
• Information Flow in Graph Neural Networks: A Clinical Triage Use Case [49.86931948849343]
  Graph Neural Networks (GNNs) have gained popularity in healthcare and other domains due to their ability to process multi-modal and multi-relational graphs. We investigate how the flow of embedding information within GNNs affects the prediction of links in Knowledge Graphs (KGs) Our results demonstrate that incorporating domain knowledge into the GNN connectivity leads to better performance than using the same connectivity as the KG or allowing unconstrained embedding propagation.
  arXiv  Detail & Related papers  (2023-09-12T09:18:12Z)
• DEGREE: Decomposition Based Explanation For Graph Neural Networks [55.38873296761104]
  We propose DEGREE to provide a faithful explanation for GNN predictions. By decomposing the information generation and aggregation mechanism of GNNs, DEGREE allows tracking the contributions of specific components of the input graph to the final prediction. We also design a subgraph level interpretation algorithm to reveal complex interactions between graph nodes that are overlooked by previous methods.
  arXiv  Detail & Related papers  (2023-05-22T10:29:52Z)
• Relation Embedding based Graph Neural Networks for Handling Heterogeneous Graph [58.99478502486377]
  We propose a simple yet efficient framework to make the homogeneous GNNs have adequate ability to handle heterogeneous graphs. Specifically, we propose Relation Embedding based Graph Neural Networks (RE-GNNs), which employ only one parameter per relation to embed the importance of edge type relations and self-loop connections.
  arXiv  Detail & Related papers  (2022-09-23T05:24:18Z)
• EvenNet: Ignoring Odd-Hop Neighbors Improves Robustness of Graph Neural Networks [51.42338058718487]
  Graph Neural Networks (GNNs) have received extensive research attention for their promising performance in graph machine learning. Existing approaches, such as GCN and GPRGNN, are not robust in the face of homophily changes on test graphs. We propose EvenNet, a spectral GNN corresponding to an even-polynomial graph filter.
  arXiv  Detail & Related papers  (2022-05-27T10:48:14Z)
• Generalizing Graph Neural Networks on Out-Of-Distribution Graphs [51.33152272781324]
  Graph Neural Networks (GNNs) are proposed without considering the distribution shifts between training and testing graphs. In such a setting, GNNs tend to exploit subtle statistical correlations existing in the training set for predictions, even though it is a spurious correlation. We propose a general causal representation framework, called StableGNN, to eliminate the impact of spurious correlations.
  arXiv  Detail & Related papers  (2021-11-20T18:57:18Z)
• Graph Backdoor [53.70971502299977]
  We present GTA, the first backdoor attack on graph neural networks (GNNs) GTA departs in significant ways: it defines triggers as specific subgraphs, including both topological structures and descriptive features. It can be instantiated for both transductive (e.g., node classification) and inductive (e.g., graph classification) tasks.
  arXiv  Detail & Related papers  (2020-06-21T19:45:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.