Graph Information Bottleneck
Tailin Wu, Hongyu Ren, Pan Li, Jure Leskovec

TL;DR
This paper introduces Graph Information Bottleneck (GIB), an information-theoretic approach to enhance the robustness of graph neural networks against adversarial attacks by balancing expressiveness and robustness.
Contribution
The paper proposes GIB, a novel regularization framework for GNNs that incorporates structural and feature information, improving adversarial robustness over existing models.
Findings
GIB models outperform state-of-the-art defenses against adversarial attacks.
GIB achieves up to 31% improvement under adversarial perturbations.
Structural and feature regularization enhances GNN robustness.
Abstract
Representation learning of graph-structured data is challenging because both graph structure and node features carry important information. Graph Neural Networks (GNNs) provide an expressive way to fuse information from network structure and node features. However, GNNs are prone to adversarial attacks. Here we introduce Graph Information Bottleneck (GIB), an information-theoretic principle that optimally balances expressiveness and robustness of the learned representation of graph-structured data. Inheriting from the general Information Bottleneck (IB), GIB aims to learn the minimal sufficient representation for a given task by maximizing the mutual information between the representation and the target, and simultaneously constraining the mutual information between the representation and the input data. Different from the general IB, GIB regularizes the structural as well as the…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
Taxonomy
TopicsAdvanced Graph Neural Networks · Adversarial Robustness in Machine Learning · Topic Modeling
