Abstract:Effective potentials are an essential ingredient of classical molecular dynamics (MD) simulations. Little is understood of the consequences of representing the complex energy landscape of an atomic configuration by an effective potential or force field containing considerably fewer parameters. The probabilistic potential ensemble method has been implemented in the potfit force matching code. This introduces uncertainty quantification into the interatomic potential generation process. Uncertainties in the effective potential are propagated through MD to obtain uncertainties in quantities of interest (QoI), which are a measure of the confidence in the model predictions. We demonstrate the technique using three potentials for nickel: two simple pair potentials, Lennard-Jones and Morse, and a local density dependent embedded atom method potential. A potential ensemble fit to density functional theory (DFT) reference data is constructed for each potential to calculate the uncertainties in lattice constants, elastic constants and thermal expansion. We quantitatively illustrate the cases of poor model selection and fit, highlighted by the uncertainties in the quantities calculated. This shows that our method can capture the effects of the error incurred in QoI resulting from the potential generation process without resorting to comparison with experiment or DFT, which is an essential part to assess the predictive power of MD simulations. Our understanding of the physics underlying material properties relies on verification from computational models of materials and molecules. Such materials simulations also allow us to predict new properties and structures which can then be reproduced experimentally. In order to facilitate the modelling, interatomic potentials have long been used to circumvent the limitations in timescale and simulation size of costly first principles calculations by specifying the energy dependence on the atomic positions. This functional representation is a key constituent of molecular dynamics (MD) simulations, where the quality of the output relies predominantly on the potential employed. Currently, the systematic error incurred in using an interatomic potential is generally unknown, as is the resulting effect on quantities of interest (QoI) it is used to predict, therein forming the motivation for this work. The general idea behind interatomic potentials is that the energy of a collection of atoms can be represented by an explicit functional form or model, dependent on the atomic separation. These analytic potential functions encode the physics into the system and contain a limited number of additional parameters which are adjusted to reproduce desired quantities. The first potentials used intuitive functional forms, fitted to experimental data; however new potentials are frequently fitted to ab-initio data such as atomic forces, energies and stresses. We use the potfit open source force matching code to fit interatomic potentials to density functional theory (DFT) data, and subsequently quantify the uncertainty incurred in simulations using the potential, in lieu of a first principles approach. There has been a significant recent effort in the literature to quantify uncertainty in this multiscale modelling process, with Bayesian frameworks proposed for a variety of interatomic models and force fields [ 1 – 3 ]. There has also been work toward the quantification of uncertainty due to the potential fitting reference set [ 4 ], as well as a proposed framework to efficiently propagate parameter uncertainties to MD outputs [ 5 ]. More specifically, quantification of parameter uncertainty for single potentials has been undertaken in a handful of cases [ 6 – 9 ]. We add to the growing body of uncertainty quantification (UQ) work in potential development and application by providing an open source implementation of the framework in [ 1 ] for use in future potential development projects. We have implemented a new module in potfit which adds a potential ensemble functionality to the potential fitting workflow. For potentials fitted with potfit , the corresponding ensemble informs of the parameter uncertainties, given the associated reference set. Section 2.1 introduces the potfit code, followed by an outline of the potential ensemble method in section 2.2 . Implementation specific details are outlined in section 3 . We then demonstrate in section 4 how an ensemble can be propagated through MD simulations, illustrating the incurred uncertainties in both equilibrium and non-equilibrium QoI: the equil -Abstract Truncated-

On the Uncertainty Estimates of Equivariant-Neural-Network-Ensembles Interatomic Potentials

Single-model uncertainty quantification in neural network potentials does not consistently outperform model ensembles

Graph Neural Network Interatomic Potential Ensembles with Calibrated Aleatoric and Epistemic Uncertainty on Energy and Forces

Evaluation of uncertainty estimations for Gaussian process regression based machine learning interatomic potentials

Improved Uncertainty Estimation of Graph Neural Network Potentials Using Engineered Latent Space Distances

Robust and scalable uncertainty estimation with conformal prediction for machine-learned interatomic potentials

Statistical methods for resolving poor uncertainty quantification in machine learning interatomic potentials

Evidential Deep Learning for Interatomic Potentials

Spatially resolved uncertainties for machine learning potentials

Efficient ensemble uncertainty estimation in Gaussian Processes Regression

Scalable Bayesian Uncertainty Quantification for Neural Network Potentials: Promise and Pitfalls

Uncertainty Estimation for Molecules: Desiderata and Methods

Uncertainty Quantification in Atomistic Simulations of Silicon using Interatomic Potentials

Transferable Water Potentials Using Equivariant Neural Networks

Evaluating Scalable Uncertainty Estimation Methods for Deep Learning-Based Molecular Property Prediction

Calibrated Uncertainty for Molecular Property Prediction using Ensembles of Message Passing Neural Networks

Discrepancies and error evaluation metrics for machine learning interatomic potentials

Uncertainty quantification for classical effective potentials: an extension to potfit

Uncertainty quantification by direct propagation of shallow ensembles

Interpolation and differentiation of alchemical degrees of freedom in machine learning interatomic potentials

Analysis of uncertainty of neural fingerprint-based models