Physical Pooling Functions in Graph Neural Networks for Molecular Property Prediction
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2022
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916090305576960 |
|---|---|
| author | Schweidtmann, Artur M. Rittig, Jan G. Weber, Jana M. Grohe, Martin Dahmen, Manuel Leonhard, Kai Mitsos, Alexander |
| author_facet | Schweidtmann, Artur M. Rittig, Jan G. Weber, Jana M. Grohe, Martin Dahmen, Manuel Leonhard, Kai Mitsos, Alexander |
| contents | Graph neural networks (GNNs) are emerging in chemical engineering for the end-to-end learning of physicochemical properties based on molecular graphs. A key element of GNNs is the pooling function which combines atom feature vectors into molecular fingerprints. Most previous works use a standard pooling function to predict a variety of properties. However, unsuitable pooling functions can lead to unphysical GNNs that poorly generalize. We compare and select meaningful GNN pooling methods based on physical knowledge about the learned properties. The impact of physical pooling functions is demonstrated with molecular properties calculated from quantum mechanical computations. We also compare our results to the recent set2set pooling approach. We recommend using sum pooling for the prediction of properties that depend on molecular size and compare pooling functions for properties that are molecular size-independent. Overall, we show that the use of physical pooling functions significantly enhances generalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2207_13779 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Physical Pooling Functions in Graph Neural Networks for Molecular Property Prediction Schweidtmann, Artur M. Rittig, Jan G. Weber, Jana M. Grohe, Martin Dahmen, Manuel Leonhard, Kai Mitsos, Alexander Machine Learning Graph neural networks (GNNs) are emerging in chemical engineering for the end-to-end learning of physicochemical properties based on molecular graphs. A key element of GNNs is the pooling function which combines atom feature vectors into molecular fingerprints. Most previous works use a standard pooling function to predict a variety of properties. However, unsuitable pooling functions can lead to unphysical GNNs that poorly generalize. We compare and select meaningful GNN pooling methods based on physical knowledge about the learned properties. The impact of physical pooling functions is demonstrated with molecular properties calculated from quantum mechanical computations. We also compare our results to the recent set2set pooling approach. We recommend using sum pooling for the prediction of properties that depend on molecular size and compare pooling functions for properties that are molecular size-independent. Overall, we show that the use of physical pooling functions significantly enhances generalization. |
| title | Physical Pooling Functions in Graph Neural Networks for Molecular Property Prediction |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2207.13779 |