Kavli Affiliate: Wei Gao | First 5 Authors: Daniel Ocampoa, Daniela Possob, Reza Namakiana, Wei Gao, | Summary: Training machine learning interatomic potentials often requires optimizing a loss function composed of three variables: potential energies, forces, and stress. The contribution of each variable to the total loss is typically weighted using fixed coefficients. Identifying these […]
Continue.. Adaptive Loss Weighting for Machine Learning Interatomic Potentials