Assessing Non-Nested Configurations of Multifidelity Machine Learning for Quantum-Chemical Properties

TL;DR

This paper evaluates the effectiveness of non-nested training data configurations in multifidelity machine learning models for quantum-chemical properties, finding that optimized MFML can perform well without nested data requirements.

Contribution

It demonstrates that the optimized MFML method can effectively handle non-nested training data, unlike traditional MFML which requires nested data structures.

Findings

MFML requires nested data for optimal performance.

o-MFML performs well with non-nested data, comparable to nested configurations.

Results are based on predicting energies of diverse molecules in the CheMFi dataset.

Abstract

Multifidelity machine learning (MFML) for quantum chemical (QC) properties has seen strong development in the recent years. The method has been shown to reduce the cost of generating training data for high-accuracy low-cost ML models. In such a set-up, the ML models are trained on molecular geometries and some property of interest computed at various computational chemistry accuracies, or fidelities. These are then combined in training the MFML models. In some multifidelity models, the training data is required to be nested, that is the same molecular geometries are included to calculate the property across all the fidelities. In these multifidelity models, the requirement of a nested configuration restricts the kind of sampling that can be performed while selection training samples at different fidelities. This work assesses the use of non-nested training data for two of these multifidelity methods, namely MFML and optimized MFML (o-MFML). The assessment is carried out for the prediction of ground state energies and first vertical excitation energies of a diverse collection of molecules of the CheMFi dataset. Results indicate that the MFML method still requires a nested structure of training data across the fidelities. However, the o-MFML method shows promising results for non-nested multifidelity training data with model errors comparable to the nested configurations.