Sparse hierarchical representation learning on molecular graphs
Abstract
Architectures for sparse hierarchical representation learning have recently been proposed for graph-structured data, but so far assume the absence of edge features in the graph. We close this gap and propose a method to pool graphs with edge features, inspired by the hierarchical nature of chemistry. In particular, we introduce two types of pooling layers compatible with an edge-feature graph-convolutional architecture and investigate their performance for molecules relevant to drug discovery on a set of two classification and two regression benchmark datasets of MoleculeNet. We find that our models significantly outperform previous benchmarks on three of the datasets and reach state-of-the-art results on the fourth benchmark, with pooling improving performance for three out of four tasks, keeping performance stable on the fourth task, and generally speeding up the training process.
- Publication:
-
arXiv e-prints
- Pub Date:
- August 2019
- DOI:
- 10.48550/arXiv.1908.02065
- arXiv:
- arXiv:1908.02065
- Bibcode:
- 2019arXiv190802065B
- Keywords:
-
- Computer Science - Machine Learning;
- Statistics - Machine Learning
- E-Print:
- 4 pages, 2 figures, accepted as a DLG 2019 workshop paper at KDD 2019