Each task probes a different aspect of machine-learning interatomic potentials, from ground-state stability prediction to finite-temperature dynamics.
Predicting ground state energies of unrelaxed structures
Relaxing atomic positions to minimize the energy
Modeling harmonic and anharmonic lattice vibrations
Predict potential energy curves for diatomic molecules.
Reproducing structural, thermodynamic and vibrational observables of ab-initio molecular dynamics trajectories