Flexible Cutoff Learning: Optimizing Machine Learning Potentials After Training
This method addresses the need for adaptable and efficient MLIPs in computational chemistry, offering a post-training optimization approach that is incremental but practical.
The paper tackles the inflexibility of machine learning interatomic potentials (MLIPs) by introducing Flexible Cutoff Learning (FCL), which allows adjusting cutoff radii after training; results show a reduction in computational cost by over 60% with less than 1% increase in force errors for molecular crystals.
We introduce Flexible Cutoff Learning (FCL), a method for training machine learning interatomic potentials (MLIPs) whose cutoff radii can be adjusted after training. Unlike conventional MLIPs that fix the cutoff radius during training, FCL models are trained by randomly sampling cutoff radii independently for each atom. The resulting model can then be deployed with different per-atom cutoff radii depending on the application, enabling application-specific optimization of the accuracy-cost tradeoff. Using a differentiable cost model, these per-atom cutoffs can be optimized for specific target systems after training. We demonstrate FCL with a modified MACE architecture trained on the MAD dataset. For a subset featuring molecular crystals, optimized per-atom cutoffs reduce computational cost by more than 60% while increasing force errors by less than 1%. These results show that FCL enables training of a single general-purpose MLIP that can be adapted to diverse applications through post-training cutoff optimization, eliminating the need for retraining.