CCGIB: A Cross-Channel Graph Information Bottleneck Principle
- 25 June 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (5) , 9488-9499
Abstract
The empirical studies of most existing graph neural networks (GNNs) broadly take the original node feature and adjacency relationship as single-channel input, ignoring the rich information of multiple graph channels. To circumvent this issue, the multichannel graph analysis framework has been developed to fuse graph information across channels. How to model and integrate shared (i.e., consistency) and channel-specific (i.e., complementarity) information is a key issue in multichannel graph analysis. In this article, we propose a cross-channel graph information bottleneck (CCGIB) principle to maximize the agreement for common representations and the disagreement for channel-specific representations. Under this principle, we formulate the consistency and complementarity information bottleneck (IB) objectives. To enable optimization, a viable approach involves deriving variational lower bound and variational upper bound (VarUB) of mutual information terms, subsequently focusing on optimizing these variational bounds to find the approximate solutions. However, obtaining the lower bounds of cross-channel mutual information objectives proves challenging through direct utilization of variational approximation, primarily due to the independence of the distributions. To address this challenge, we leverage the inherent property of joint distributions and subsequently derive variational bounds to effectively optimize these information objectives. Extensive experiments on graph benchmark datasets demonstrate the superior effectiveness of the proposed method.Keywords
Funding Information
- National Natural Science Foundation of China (62206208, 62306226)
- Fundamental Research Funds for the Central Universities (XJSJ23021)
- Key Research and Development Project of Shaanxi Province (2022QCY-LL-35)
This publication has 28 references indexed in Scilit:
- Learnable Graph Convolutional Network With Semisupervised Graph Information BottleneckIEEE Transactions on Neural Networks and Learning Systems, 2023
- Unpaired Multi-View Graph Clustering With Cross-View Structure MatchingIEEE Transactions on Neural Networks and Learning Systems, 2023
- Efficient Multi-View Clustering via Unified and Discrete Bipartite Graph LearningIEEE Transactions on Neural Networks and Learning Systems, 2023
- Multi-View Graph Learning by Joint Modeling of Consistency and InconsistencyIEEE Transactions on Neural Networks and Learning Systems, 2022
- Propagation Enhanced Neural Message Passing for Graph Representation LearningIEEE Transactions on Knowledge and Data Engineering, 2021
- Multi-View Clustering via Nonnegative and Orthogonal Graph ReconstructionIEEE Transactions on Neural Networks and Learning Systems, 2021
- Deep Neural Message Passing With Hierarchical Layer Aggregation and Neighbor NormalizationIEEE Transactions on Neural Networks and Learning Systems, 2021
- The Graph Neural Network ModelIEEE Transactions on Neural Networks, 2008
- Weighted Graph Cuts without Eigenvectors A Multilevel ApproachIEEE Transactions on Pattern Analysis and Machine Intelligence, 2007
- Protein function prediction via graph kernelsBioinformatics, 2005