Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck
This study aims to build a pre-trained Graph Neural Network (GNN) model on molecules without human annotations or prior knowledge. Although various attempts have been proposed to overcome limitations in acquiring labeled molecules, the previous pre-training methods still rely on semantic subgraphs, i.e., functional groups. Only focusing on the functional groups could overlook the graph-level distinctions. The key challenge to build a pre-trained GNN on molecules is how to (1) generate well-distinguished graph-level representations and (2) automatically discover the functional groups without prior knowledge. To solve it, we propose a novel Subgraph-conditioned Graph Information Bottleneck, named S-CGIB, for pre-training GNNs to recognize core subgraphs (graph cores) and significant subgraphs. The main idea is that the graph cores contain compressed and sufficient information that could generate well-distinguished graph-level representations and reconstruct the input graph conditioned on significant subgraphs across molecules under the S-CGIB principle. To discover significant subgraphs without prior knowledge about functional groups, we propose generating a set of functional group candidates, i.e., ego networks, and using an attention-based interaction between the graph core and the candidates. Despite being identified from self-supervised learning, our learned subgraphs match the real-world functional groups. Extensive experiments on molecule datasets across various domains demonstrate the superiority of S-CGIB.
Code (1)
Tasks
Graph ClassificationGraph Neural NetworkGraph RegressionMolecular Property PredictionSelf-Supervised LearningMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck
This study aims to build a pre-trained Graph Neural Network (GNN) model on molecules without human annotations or prior knowledge. Although various attempts have been proposed to overcome limitations in acquiring labeled…
Graph Neural NetworkSelf-Supervised LearningPredicting Drug Interactions and Mutagenicity with Ensemble Classifiers on Subgraphs of Molecules
In this study, we intend to solve a mutual information problem in interacting molecules of any type, such as proteins, nucleic acids, and small molecules. Using machine learning techniques, we accurately predict pairwise…
BIG-bench Machine LearningMotif-Driven Contrastive Learning of Graph Representations
Pre-training Graph Neural Networks (GNN) via self-supervised contrastive learning has recently drawn lots of attention. However, most existing works focus on node-level contrastive learning, which cannot capture global g…
ClusteringContrastive LearningSelf-Supervised LearningAtomic and Subgraph-aware Bilateral Aggregation for Molecular Representation Learning
Molecular representation learning is a crucial task in predicting molecular properties. Molecules are often modeled as graphs where atoms and chemical bonds are represented as nodes and edges, respectively, and Graph Neu…
Molecular Property Predictionmolecular representationProperty PredictionRepresentation Learning+1Molecular Graph Generation by Decomposition and Reassembling
Designing molecular structures with desired chemical properties is an essential task in drug discovery and material design. However, finding molecules with the optimized desired properties is still a challenging task due…
Drug DiscoveryGraph GenerationMolecular Graph Generationvalid