Machine learning predicts when to trust density functional theory results
Risk-Controlled DFT Selection for Machine-Learned Interatomic Potentials
Machine Learning
Summary
Calculating material properties using density functional theory (DFT) is slow and often the bottleneck in discovering new materials. Machine-learned models (MLIPs) can predict these properties much faster but sometimes make big errors. The authors created a method that estimates the risk of trusting the MLIP’s predictions and decides when to do a slow DFT calculation instead. This approach reduces the number of expensive DFT calculations while keeping the overall error below a user-set level.
What this means in practice
- •For materials simulation teams: Allocate expensive DFT calculations selectively based on predicted MLIP reliability to speed up materials discovery workflows.
- •For chemical informatics engineers: Improve adaptive sampling strategies by predicting when to trust fast ML models or fallback to slower quantum calculations.
Authors
Kun Ryu, Yeo Jin Jung, Shyue Ping Ong, Claire Donnat, Ying Shirley Meng
Abstract
Automated materials discovery requires evaluating large pools of candidate configurations, with density functional theory (DFT) as a major throughput bottleneck. Pretrained machine-learned interatomic potentials (MLIPs) approximate DFT cheaply, but their errors vary by orders of magnitude across configurations, making it difficult to decide when to trust the MLIP predictions and when to fall back to DFT. We frame this decision as a risk-controlled selection problem and solve it within a conformal risk control (CRC) framework. Each configuration is mapped from atomic numbers and Cartesian coordinates to a fixed geometric descriptor, which a lightweight predictor uses to estimate an MLIP-DFT disagreement score in energy, average force, and maximum force. CRC then calibrates this score into a selection threshold for DFT labeling such that the MLIP error on the unselected configurations, which keep their MLIP labels, stays below a user-chosen fraction $α$ of the expected error. Under an identical guarantee across eight dataset-MLIP pairs, our predictor routes 7% and 14% fewer configurations to DFT at $α=0.3$ than committee baselines. Because committee disagreement overlooks errors the MLIPs share, our gains are largest when the DFT labels and MLIP training data differ in DFT functional and basis set, and narrow when the two agree. Without model access, the structure-only predictor also ranks errors as well as the target MLIP's latent representation and outperforms widely used descriptors such as Smooth Overlap of Atomic Positions (SOAP) and Atom-Centered Symmetry Functions (ACSF).