CrysXPP:An Explainable Property Predictor for Crystalline Materials

TL;DR

CrysXPP is an explainable deep learning framework that predicts various properties of crystalline materials efficiently, requiring less labeled data and outperforming traditional methods and recent baselines.

Contribution

It introduces CrysAE, an autoencoder trained on untagged data, enabling effective property prediction with minimal labeled data and providing interpretability.

Findings

Outperforms recent baseline algorithms across seven properties.

Requires significantly less tagged data than traditional models.

Outperforms conventional DFT with limited experimental data.

Abstract

We present a deep-learning framework, CrysXPP, to allow rapid prediction of electronic, magnetic and elastic properties of a wide range of materials with reasonable precision. Although our work is consistent with several recent attempts to build deep learning-based property predictors, it overcomes some of their limitations. CrysXPP lowers the need for a large volume of tagged data to train a deep learning model by intelligently designing an autoencoder CrysAE and passing the structural information to the property prediction process. The autoencoder in turn is trained on a huge volume of untagged crystal graphs, the designed loss function helps in capturing all their important structural and chemical information. Moreover, CrysXPP uses only a small amount of tagged data for property prediction, and also trains a feature selector that provides interpretability to the results obtained. We demonstrate that CrysXPP convincingly performs better than all the competing and recent baseline algorithms across seven diverse set of properties. Most notably, when given a small amount of experimental data, CrysXPP is consistently able to outperform conventional DFT. We release the large pretrained model CrysAE so that it could be fine-tuned using small amount of tagged data by the research community on various applications with restricted data source.