Not from Scratch: Predicting Thermophysical Properties through Model-Based Transfer Learning Using Graph Convolutional Networks

Rodrigo S. Hormazabal, Jeong Won Kang, Kiho Park, Dae Ryook Yang

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


In this study, a framework for the prediction of thermophysical properties based on transfer learning from existing estimation models is explored. The predictive capabilities of conventional group-contribution methods and traditional machine-learning approaches rely heavily on the availability of experimental datasets and their uncertainty. Through the use of a pretraining scheme, which leverages the knowledge established by other estimation methods, improved prediction models for thermophysical properties can be obtained after fine-tuning networks with more accurate experimental data. As our experiments show, for the case of critical properties of compounds, this pipeline not only improves the performance of the models on commonly found organic structures but can also help these models generalize to less explored areas of chemical space, where experimental data is scarce, such as inorganics and heavier organic compounds. Transfer learning from estimation models data also allows for graph-based deep learning models to create more flexible molecular features over a bigger chemical space, which leads to improved predictive capabilities and can give insights into the relationship between molecular structures and thermophysical properties. The generated molecular features can discriminate behavior discrepancy between isomers without the need of additional parameters. Also, this approach shows better robustness to outliers in experimental datasets.

Original languageEnglish
Pages (from-to)5411-5424
Number of pages14
JournalJournal of Chemical Information and Modeling
Issue number22
Publication statusPublished - 2022 Nov 28

Bibliographical note

Funding Information:
This work is supported by the Korea University.

Publisher Copyright:
© 2022 American Chemical Society. All rights reserved.

ASJC Scopus subject areas

  • General Chemistry
  • General Chemical Engineering
  • Computer Science Applications
  • Library and Information Sciences


Dive into the research topics of 'Not from Scratch: Predicting Thermophysical Properties through Model-Based Transfer Learning Using Graph Convolutional Networks'. Together they form a unique fingerprint.

Cite this