Bias in Universal Machine-Learned Interatomic Potentials and Its Effects on Fine-Tuning
- 1Department of Chemical Engineering
- 2Georgia Institute of Technology
Abstract
Universal machine learned interatomic potentials (uMLIPs) embody a growing area of interest due to their transferability across the periodic table, displaying an error of about 0.6 kcal/mol against the Matbench Discovery test set. However, we show that achieving more accurate predictions on out-of-domain tasks requires fine-tuning. Additionally, we investigate the existence and influence of model biases in molecular dynamics (MD) by examining two approaches for data generation: from multiple MD trajectories in parallel, which we call naive fine-tuning, and from a single MD trajectory with fine-tuning after set intervals, which we call iterative fine-tuning. Our results find that naive fine-tuning generates constrained data sets that fail to represent MD simulations, and thus downstream fine-tuned models fail during extrapolation. In contrast, iterative fine-tuning yields models that are more generalizable and accurate, producing stable dynamics. These findings indicate the role of uMLIP bias in fine-tuning, and highlight the need for multiple fine-tuning steps. Lastly, we relate unphysical behavior to principal component space, and quantify extrapolations through Q-residual analysis, which are useful as a proxy for epistemic uncertainty for larger simulations.
Related
- Correcting the Liquid Densities of Universal Machine-Learning Interatomic Potentials Using Local-Number-Density Repulsion
- Universal Interatomic Potentials as Configuration-Space Generators for One-Shot and Iterative Fine-Tuning of Ab Initio-Accurate Material-Specific Models
- Reference Conventions and Structure Types, Rather Than 4f Occupancy, Shape Lanthanide Errors in Universal Machine-Learning Interatomic Potentials
- Neural Constrained Charge Equilibration for Extracting Charge Transport Parameters from Machine Learned Interatomic Potentials
- FACTO: fast atom-centered tensor operator for accurate and efficient foundational machine-learned interatomic potentials
BibTeX
@article{wong2026bias,
title = {{Bias in Universal Machine-Learned Interatomic Potentials and Its Effects on Fine-Tuning}},
author = {Nicolas Wong and Julia H. Yang},
journal = {Journal of Chemical Theory and Computation},
year = {2026},
month = jun,
doi = {10.1021/acs.jctc.6c00425},
url = {https://doi.org/10.1021/acs.jctc.6c00425},
note = {CC BY},
}
RIS
TY - JOUR
TI - Bias in Universal Machine-Learned Interatomic Potentials and Its Effects on Fine-Tuning
AU - Nicolas Wong
AU - Julia H. Yang
PY - 2026
DA - 2026/06/20
JO - Journal of Chemical Theory and Computation
DO - 10.1021/acs.jctc.6c00425
UR - https://doi.org/10.1021/acs.jctc.6c00425
AB - Universal machine learned interatomic potentials (uMLIPs) embody a growing area of interest due to their transferability across the periodic table, displaying an error of about 0.6 kcal/mol against the Matbench Discovery test set. However, we show that achieving more accurate predictions on out-of-domain tasks requires fine-tuning. Additionally, we investigate the existence and influence of model biases in molecular dynamics (MD) by examining two approaches for data generation: from multiple MD trajectories in parallel, which we call naive fine-tuning, and from a single MD trajectory with fine-tuning after set intervals, which we call iterative fine-tuning. Our results find that naive fine-tuning generates constrained data sets that fail to represent MD simulations, and thus downstream fine-tuned models fail during extrapolation. In contrast, iterative fine-tuning yields models that are more generalizable and accurate, producing stable dynamics. These findings indicate the role of uMLIP bias in fine-tuning, and highlight the need for multiple fine-tuning steps. Lastly, we relate unphysical behavior to principal component space, and quantify extrapolations through Q-residual analysis, which are useful as a proxy for epistemic uncertainty for larger simulations.
SN - 1549-9618
SN - 1549-9626
ER -