References# Adversarial methods# FGSM: Explaining and Harnessing Adversarial Examples I-FGSM: Adversarial Examples in the Physical World PGD: Towards Deep Learning Models Resistant to Adversarial Attacks Machine learning force fields# MACE architecture paper MACE-MH paper MACE-MH model documentation UMA paper UMA model documentation CHGNet documentation LiCOHPF database with pre-trained MACE + MTP Validation and failure analysis# How to validate machine-learned interatomic potentials MLFF error evaluation MLFFs for modelling deformation MLFFs for material properties Correct energies does not guarantee the correct trajectory Fine-tuning machine-learned interatomic potentials Variance in machine learning force fields Structural metrics# NumPy vector-norm documentation Jaccard-distance reference A Note on the Triangle Inequality for the Jaccard Distance SciPy Jaccard-distance documentation Radial distribution functions ASE RDF implementation Coordination-number analysis ASE neighbour-list documentation Pymatgen symmetry-analysis documentation Spglib documentation NumPy quantile documentation Scikit-learn coefficient-of-determination documentation Simulation and structure tools# MLFF Attack source repository Atomic Simulation Environment documentation ASE geometry-optimization documentation ASE neighbour-list, natural-cutoff, and connectivity documentation ASE supercell implementation Contour exploration# ASE contour exploration documentation Contour exploration method