paper-with-me

홈 › Papers

Graph Information Bottleneck

2020-10-24 · NeurIPS 2020 12 · Tailin Wu, Hongyu Ren, Pan Li, Jure Leskovec

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 feature information. We design two sampling algorithms for structural regularization and instantiate the GIB principle with two new models: GIB-Cat and GIB-Bern, and demonstrate the benefits by evaluating the resilience to adversarial attacks. We show that our proposed models are more robust than state-of-the-art graph defense models. GIB-based models empirically achieve up to 31% improvement with adversarial perturbation of the graph structure as well as node features.

📄 PDF Abstract BibTeX arXiv:2010.12811

Code (1)

snap-stanford/GIB 공식 구현 pytorch

Tasks

Representation Learning

Similar Papers 제목 키워드 기반

Improving Subgraph Recognition with Variational Graph Information Bottleneck

2021-12-18 · CVPR 2022 1 · Junchi Yu, Jie Cao, Ran He

Subgraph recognition aims at discovering a compressed substructure of a graph that is most informative to the graph property. It can be formulated by optimizing Graph Information Bottleneck (GIB) with a mutual informatio…

Graph Classification

Individual and Structural Graph Information Bottlenecks for Out-of-Distribution Generalization

2023-06-28 · Ling Yang, Jiayi Zheng, Heyuan Wang, Zhongyi Liu 외

Out-of-distribution (OOD) graph generalization are critical for many real-world applications. Existing methods neglect to discard spurious or noisy features of inputs, which are irrelevant to the label. Besides, they mai…

Graph LearningOut-of-Distribution Generalization

Recognizing Predictive Substructures with Subgraph Information Bottleneck

2021-03-20 · Junchi Yu, Tingyang Xu, Yu Rong, Yatao Bian 외

The emergence of Graph Convolutional Network (GCN) has greatly boosted the progress of graph learning. However, two disturbing factors, noise and redundancy in graph data, and lack of interpretation for prediction result…

Bilevel OptimizationGraph Learning

Graph Structure Learning with Variational Information Bottleneck

2021-12-16 · Qingyun Sun, JianXin Li, Hao Peng, Jia Wu 외

Graph Neural Networks (GNNs) have shown promising results on a broad spectrum of applications. Most empirical studies of GNNs directly take the observed graph as input, assuming the observed structure perfectly depicts t…

Graph structure learning

Towards Consistency and Complementarity: A Multiview Graph Information Bottleneck Approach

2022-10-11 · Xiaolong Fan, Maoguo Gong, Yue Wu, Mingyang Zhang 외

The empirical studies of Graph Neural Networks (GNNs) broadly take the original node feature and adjacency relationship as singleview input, ignoring the rich information of multiple graph views. To circumvent this issue…