On Graph Neural Network Ensembles for Large-Scale Molecular Property Prediction
arXiv:2106.15529
Abstract
In order to advance large-scale graph machine learning, the Open Graph Benchmark Large Scale Challenge (OGB-LSC) was proposed at the KDD Cup 2021. The PCQM4M-LSC dataset defines a molecular HOMO-LUMO property prediction task on about 3.8M graphs. In this short paper, we show our current work-in-progress solution which builds an ensemble of three graph neural networks models based on GIN, Bayesian Neural Networks and DiffPool. Our approach outperforms the provided baseline by 7.6%. Moreover, using uncertainty in our ensemble's prediction, we can identify molecules whose HOMO-LUMO gaps are harder to predict (with Pearson's correlation of 0.5181). We anticipate that this will facilitate active learning.
arXiv comment7 pages, 1 figure, 1 table
References in corpus (8)
- Neural Message Passing for Quantum Chemistry
- Weight Uncertainty in Neural Networks
- Fast Graph Representation Learning with PyTorch Geometric
- Hierarchical Graph Representation Learning with Differentiable Pooling
- Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
- Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
- OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphs
- Bayesian Graph Neural Networks for Molecular Property Prediction