Few-Shot Graph Learning for Molecular Property Prediction

Abstract

The recent success of graph neural networks has significantly boosted molecular property prediction, advancing activities such as drug discovery. The existing deep neural network methods usually require large training dataset for each property, impairing their performances in cases (especially for new molecular properties) with a limited amount of experimental data, which are common in real situations. To this end, we propose Meta-MGNN, a novel model for few-shot molecular property prediction. Meta-MGNN applies molecular graph neural network to learn molecular representation and builds a meta-learning framework for model optimization. To exploit unlabeled molecular information and address task heterogeneity of different molecular properties, Meta-MGNN further incorporates molecular structure, attribute based self-supervised modules and self-attentive task weights into the former framework, strengthening the whole learning model. Extensive experiments on two public multi-property datasets demonstrate that Meta-MGNN outperforms a variety of state-of-the-art methods.

arXiv commentTo appear in WWW 2021 (long paper); Code is available at https://github.com/zhichunguo/Meta-MGNN

References in corpus (12)

• Distributed Representations of Words and Phrases and their Compositionality
• Semi-Supervised Classification with Graph Convolutional Networks
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• A Structured Self-attentive Sentence Embedding
• Session-based Social Recommendation via Dynamic Graph Attention Networks
• Few-Shot Learning with Graph Neural Networks
• Strategies for Pre-training Graph Neural Networks
• Molecular geometry prediction using a deep generative graph neural network
• Meta-GNN: On Few-shot Node Classification in Graph Meta-learning
• Action Sequence Augmentation for Early Graph-based Anomaly Detection

Cited by in corpus (15)

• A Comprehensive Survey on Deep Graph Representation Learning
• Advances of Machine Learning in Materials Science: Ideas and Techniques
• Graph Rationalization with Environment-based Augmentations
• Condensing Graphs via One-Step Gradient Matching
• Feature Overcorrelation in Deep Graph Neural Networks: A New Perspective
• Action Sequence Augmentation for Early Graph-based Anomaly Detection
• Data-Efficient Brain Connectome Analysis via Multi-Task Meta-Learning
• Topology-aware Embedding Memory for Continual Learning on Expanding Networks
• Meta-Learning with Graph Neural Networks: Methods and Applications
• Multi-View Subgraph Neural Networks: Self-Supervised Learning with Scarce Labeled Data
• Improving Compound Activity Classification via Deep Transfer and Representation Learning
• Contrastive Dual-Interaction Graph Neural Network for Molecular Property Prediction
• CARL-G: Clustering-Accelerated Representation Learning on Graphs
• Multi-Level Fusion Graph Neural Network for Molecule Property Prediction
• PACIA: Parameter-Efficient Adapter for Few-Shot Molecular Property Prediction