Force-Aware Neural Tangent Kernels for Scalable and Robust Active Learning of MLIPs
Not on the map. Why?
Abstract (arXiv)
Active learning for machine-learning interatomic potentials (MLIPs) must address several challenges to be practical: scaling to large candidate pools, leveraging energy-force supervision, and maintaining robustness when candidate pools are biased relative to the target distribution. In this work, we jointly address these challenges. We first introduce a linearly scaling acquisition framework based on chunked feature-space posterior-variance shortlisting. By avoiding materialisation of the candidate and train set kernels, this approach enables screening of ~200k structures within hours and applies broadly to acquisition strategies that score candidates based on molecular similarity metrics. We then extend the Neural Tangent Kernel (NTK) to a force-aware setting via mixed parameter-coordinate derivatives, yielding a force NTK and a joint energy-force NTK that provide natural similarity metrics for vector-field prediction. We demonstrate the effectiveness of the joint energy-force NTK on the OC20 dataset, where force-aware acquisition is crucial: it achieves the lowest energy and force MAE and RMSE across all metrics and distribution splits. Across T1x, PMechDB, and RGD benchmarks, our force NTK methods remain competitive with established baselines while being significantly more efficient than committee-based approaches. Under a controlled candidate-pool shift case study on T1x, acquisition based on pretrained MLIP embeddings and NTKs remains robust, whereas committee-based methods exhibit higher variance. Overall, these results show that a single pretrained MLIP can enable scalable, force-aware, and distribution-robust active learning for foundation-model fine-tuning.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 15 of its 51 references that are in the map. First posted on arXiv 13 May 2026.
Builds on
15 of 51 references.
References · 15
- 2018Neural Tangent Kernel: Convergence and Generalization in Neural Networkscited by 2,621
- 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 708
- 2022MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fieldscited by 474
- 2020The Open Catalyst 2020 (OC20) Dataset and Community Challengescited by 246
- 2018Less is more: sampling chemical space with active learningcited by 82
- 2022SPICE, A Dataset of Drug-like Molecules and Peptides for Training Machine Learning Potentialscited by 82
- 2019DScribe: Library of Descriptors for Machine Learning in Materials Sciencecited by 63
- 2022A Framework and Benchmark for Deep Batch Active Learning for Regressioncited by 49
- 2018Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulationcited by 37
- 2022Transition1x -- a Dataset for Building Generalizable Reactive Machine Learning Potentialscited by 37
- 2025A practical guide to machine learning interatomic potentials -- Status and futurecited by 26
- 2023Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic potentialscited by 16
3 more
- 2020Committee neural network potentials control generalization errors and enable active learningcited by 15
- 2026Data Curation for Machine Learning Interatomic Potentials by Determinantal Point Processescited by 3
- 2026Pretrained Model Representations as Acquisition Signals for Active Learning of MLIPscited by 3
Built on by
Citing papers per year: 2026: 2.
2 papers.