Issue 28, 2024

Identifying and embedding transferability in data-driven representations of chemical space

Abstract

Transferability, especially in the context of model generalization, is a paradigm of all scientific disciplines. However, the rapid advancement of machine learned model development threatens this paradigm, as it can be difficult to understand how transferability is embedded (or missed) in complex models developed using large training data sets. Two related open problems are how to identify, without relying on human intuition, what makes training data transferable; and how to embed transferability into training data. To solve both problems for ab initio chemical modelling, an indispensable tool in everyday chemistry research, we introduce a transferability assessment tool (TAT) and demonstrate it on a controllable data-driven model for developing density functional approximations (DFAs). We reveal that human intuition in the curation of training data introduces chemical biases that can hamper the transferability of data-driven DFAs. We use our TAT to motivate three transferability principles; one of which introduces the key concept of transferable diversity. Finally, we propose data curation strategies for general-purpose machine learning models in chemistry that identify and embed the transferability principles.

Graphical abstract: Identifying and embedding transferability in data-driven representations of chemical space

Supplementary files

Transparent peer review

To support increased transparency, we offer authors the option to publish the peer review history alongside their article.

View this article’s peer review history

Article information

Article type
Edge Article
Submitted
10 Apr 2024
Accepted
02 Jun 2024
First published
21 Jun 2024
This article is Open Access

All publication charges for this article have been paid for by the Royal Society of Chemistry
Creative Commons BY license

Chem. Sci., 2024,15, 11122-11133

Identifying and embedding transferability in data-driven representations of chemical space

T. Gould, B. Chan, S. G. Dale and S. Vuckovic, Chem. Sci., 2024, 15, 11122 DOI: 10.1039/D4SC02358G

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. You can use material from this article in other publications without requesting further permissions from the RSC, provided that the correct acknowledgement is given.

Read more about how to correctly acknowledge RSC content.

Social activity

Spotlight

Advertisements