Evidential Deep Learning for Interatomic Potential

<b>Dataset Description</b>This repository contains checkpoints and simulation trajectories from the paper <i>"Evidential Deep Learning for Interatomic Potential."</i><b>ckpt.zip</b>: Checkpoints for the eIP model trained on multiple datasets. For the silica glass dataset, additional checkpoints trained using ensemble, Monte Carlo dropout, Gaussian Mixture Model (GMM), and Maximum Variance Estimation (MVE) are provided.<b>traj.zip</b>: Molecular dynamics (MD) and uncertainty-driven dynamics (UDD) simulation trajectories for water, lithium iron phosphate (LiFePO₄), and polydimethylsiloxane (PDMS). All trajectories are saved in the extended XYZ format.<br>