Transfer learning for atomistic simulations using GNNs and kernel mean embeddings

TL;DR

This paper introduces a transfer learning approach combining GNNs and kernel mean embeddings to improve atomistic potential energy surface modeling, reducing data requirements and enhancing transferability.

Contribution

It presents a novel transfer learning algorithm that leverages pre-trained GNNs and kernel methods, incorporating chemical species information for better accuracy and interpretability.

Findings

Excellent generalization and transferability demonstrated on complex datasets

Outperforms methods using only GNNs or ridge regression

Improves upon existing fine-tuning approaches

Abstract

Interatomic potentials learned using machine learning methods have been successfully applied to atomistic simulations. However, accurate models require large training datasets, while generating reference calculations is computationally demanding. To bypass this difficulty, we propose a transfer learning algorithm that leverages the ability of graph neural networks (GNNs) to represent chemical environments together with kernel mean embeddings. We extract a feature map from GNNs pre-trained on the OC20 dataset and use it to learn the potential energy surface from system-specific datasets of catalytic processes. Our method is further enhanced by incorporating into the kernel the chemical species information, resulting in improved performance and interpretability. We test our approach on a series of realistic datasets of increasing complexity, showing excellent generalization and transferability performance, and improving on methods that rely on GNNs or ridge regression alone, as well as similar fine-tuning approaches.