3-D Inorganic Crystal Structure Generation and Property Prediction via Representation Learning
Published in Journal of Chemical Information and Modeling, 2020
Recommended citation: Court C.J, Yildirim B., Jain A. & Cole J.M "3-D Inorganic Crystal Structure Generation and Property Prediction via Representation Learning" Journal of Chemical Information and Modeling (accepted for publication) (2020) https://pubs.acs.org/doi/10.1021/acs.jcim.0c00464
Description
This work presents a complete representation learning pipeline for generating novel 3-d crystal structures and predicting their DFT properties. This system comprises a Conditional Deep Feature Consistent Variational Autoencoder, Unet semantic segmentation network and a crystal graph neural network.
The results show that it is possible to create highly stable and realistic crystal structures. You can check out the source code here