Magnetic interatomic potentials need to account for coupled lattice and spin degrees of freedom, yet constructing reliable training sets remains costly because noncollinear first-principles labels are expensive. Active learning can mitigate this cost, provided that the uncertainty estimate is physically meaningful for the magnetic-response targets that drive spin reorientation. Here we introduce equivariant evidential deep learning for spin-lattice potentials $\left({\mathrm{e}}^2\mathrm{SLP}\right)$, a magnetic uncertainty-learning method that formulates the projected spin-force likelihood and the corresponding epistemic uncertainty in the tangent plane orthogonal to the local spin direction. This geometry is consistent with the constrained-moment labels, which contain no independent radial response, and yields a structure-level projected-spin-force uncertainty score for candidate selection. Using bulk ${\mathrm{BiFeO}}_3$ and monolayer ${\mathrm{CrTe}}_2$ as benchmark systems, we show that this score correlates strongly with prediction errors and that the full ${\mathrm{e}}^2\mathrm{SLP}$ selection, training, and fine-tuning workflow lowers energy, force, and projected spin-force errors relative to random sampling. For monolayer ${\mathrm{CrTe}}_2$, round-by-round comparisons show accuracy competitive with a four-model committee, while a four-seed ablation separates the contributions of uncertainty-guided selection and ${\mathrm{e}}^2\mathrm{SLP}$ training followed by fine tuning. ${\mathrm{e}}^2\mathrm{SLP}$ thus provides a single-model tangent-plane uncertainty criterion for data-efficient construction of magnetic machine-learning interatomic potentials.

Close

PrevNext

View figure in article

• Interatomic & molecular potentials
• Spin dynamics
• Machine learning