EL-MLFFs: Ensemble Learning of Machine Leaning Force Fields
This work addresses the problem of model selection and force prediction accuracy in MLFFs for molecular simulations, offering a promising solution that is incremental in nature.
The paper tackles the challenge of accurately predicting atomic forces in machine learning force fields (MLFFs) by proposing an ensemble learning framework called EL-MLFFs, which integrates predictions from diverse MLFFs using a stacking method with a graph neural network as the meta-model, resulting in significantly improved force prediction accuracy on datasets like methane molecules and methanol adsorbed on a Cu(100) surface, with the ensemble of all eight models yielding the best performance.
Machine learning force fields (MLFFs) have emerged as a promising approach to bridge the accuracy of quantum mechanical methods and the efficiency of classical force fields. However, the abundance of MLFF models and the challenge of accurately predicting atomic forces pose significant obstacles in their practical application. In this paper, we propose a novel ensemble learning framework, EL-MLFFs, which leverages the stacking method to integrate predictions from diverse MLFFs and enhance force prediction accuracy. By constructing a graph representation of molecular structures and employing a graph neural network (GNN) as the meta-model, EL-MLFFs effectively captures atomic interactions and refines force predictions. We evaluate our approach on two distinct datasets: methane molecules and methanol adsorbed on a Cu(100) surface. The results demonstrate that EL-MLFFs significantly improves force prediction accuracy compared to individual MLFFs, with the ensemble of all eight models yielding the best performance. Moreover, our ablation study highlights the crucial roles of the residual network and graph attention layers in the model's architecture. The EL-MLFFs framework offers a promising solution to the challenges of model selection and force prediction accuracy in MLFFs, paving the way for more reliable and efficient molecular simulations.