Better without U: Impact of Selective Hubbard U Correction on Foundational MLIPs

• URL: http://arxiv.org/abs/2601.21056v1
• Date: Wed, 28 Jan 2026 21:24:04 GMT
• Title: Better without U: Impact of Selective Hubbard U Correction on Foundational MLIPs
• Authors: Thomas Warford, Fabian L. Thiemann, Gábor Csányi,
• Abstract summary: We show inconsistencies from the Materials Project's selective use of the Hubbard U correction, which is applied to certain transition metals only if O or F atoms are present in the simulation cell.<n>This inconsistent use of +U creates two potential-energy surfaces (PES): a lower-energy GGA surface and a higher-energy GGA+U one.<n>When trained on both, MLIPs interpolate between them, leading to systematic underbinding, or even spurious repulsion between U-corrected metals and oxygen- or fluorine-containing species.
• Score: 2.24303609250571
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The training of foundational machine learning interatomic potentials (fMLIPs) relies on diverse databases with energies and forces calculated using ab initio methods. We show that fMLIPs trained on large datasets such as MPtrj, Alexandria, and OMat24 encode inconsistencies from the Materials Project's selective use of the Hubbard U correction, which is applied to certain transition metals only if O or F atoms are present in the simulation cell. This inconsistent use of +U creates two incompatible potential-energy surfaces (PES): a lower-energy GGA surface and a higher-energy GGA+U one. When trained on both, MLIPs interpolate between them, leading to systematic underbinding, or even spurious repulsion, between U-corrected metals and oxygen- or fluorine-containing species. Models such as MACE-OMAT and -MPA exhibit repulsion between U-corrected metals and their oxides, limiting their value for studying catalysis and oxidation. We link the severity of this pathology to the oxygen number density in U-corrected training configurations. This explains why OMAT-trained models are most affected and suggests the issue might worsen as expanding future datasets increasingly include configurations with low oxygen content, such as those generated through combinatorial exploration of multi-element or defect-containing systems. Our simple per-U-corrected-atom shift aligns PBE+U and PBE energies for identical structures, yielding a smoother PES compared to existing correction schemes, which target phase diagram accuracy. As a result, models trained on datasets with our shift applied exhibit smaller mean absolute errors for the adsorption energies of oxygen on U-corrected elemental slabs. Since datasets omitting +U entirely (e.g. MatPES, MP-ALOE) avoid these pathologies, we recommend excluding +U in future fMLIP datasets. For existing datasets, our post-hoc correction provides a low-cost improvement.

Related papers

• ADAPT: Lightweight, Long-Range Machine Learning Force Fields Without Graphs [6.683253616106448]
  First-principles methods are widely used to compute defect energetics and structures, including at scale for defect databases.<n>MLFFs are computationally expensive, making machine-learning force fields (MLFFs) an attractive alternative for accelerating structural relaxations.<n>Most existing MLFFs are based on graph neural networks (GNNs), which can suffer from oversmoothing and poor representation of long-range interactions.<n>We introduce the Accelerated Deep Atomic Potential Transformer (ADAPT), an MLFF that replaces graph representations with a direct coordinates-in-space formulation.
  arXiv  Detail & Related papers  (2025-09-28T23:18:57Z)
• Learning Inter-Atomic Potentials without Explicit Equivariance [24.438029202222555]
  We introduce TransIP: Transformer-based Inter-Atomic Potentials, a novel training paradigm for interatomic potentials.<n>Our approach guides a generic non-equivariant Transformer-based model to learn SO(3)-equivariance by optimizing its representations in the embedding space.<n>Compared to a data augmentation baseline, TransIP achieves 40% to 60% improvement in performance across varying OMol25 dataset sizes.
  arXiv  Detail & Related papers  (2025-09-25T22:15:10Z)
• Efficient Federated Learning with Heterogeneous Data and Adaptive Dropout [62.73150122809138]
  Federated Learning (FL) is a promising distributed machine learning approach that enables collaborative training of a global model using multiple edge devices.<n>We propose the FedDHAD FL framework, which comes with two novel methods: Dynamic Heterogeneous model aggregation (FedDH) and Adaptive Dropout (FedAD)<n>The combination of these two methods makes FedDHAD significantly outperform state-of-the-art solutions in terms of accuracy (up to 6.7% higher), efficiency (up to 2.02 times faster), and cost (up to 15.0% smaller)
  arXiv  Detail & Related papers  (2025-07-14T16:19:00Z)
• Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion [55.95767828747407]
  In domains such as molecular and protein generation, physical systems exhibit inherent symmetries that are critical to model.<n>We present a framework that reduces training variance and provides a provably lower-variance gradient estimator.<n>We also present a practical implementation of this estimator incorporating the loss and sampling procedure through a method we call Orbit Diffusion.
  arXiv  Detail & Related papers  (2025-02-14T03:26:57Z)
• Overcoming systematic softening in universal machine learning interatomic potentials by fine-tuning [3.321322648845526]
  Machine learning interatomic potentials (MLIPs) have introduced a new paradigm for atomic simulations. Recent advancements have seen the emergence of universal MLIPs (uMLIPs) that are pre-trained on diverse materials datasets. We show that their performance in extrapolating to out-of-distribution complex atomic environments remains unclear.
  arXiv  Detail & Related papers  (2024-05-11T22:30:47Z)
• Machine Learning Force Fields with Data Cost Aware Training [94.78998399180519]
  Machine learning force fields (MLFF) have been proposed to accelerate molecular dynamics (MD) simulation. Even for the most data-efficient MLFFs, reaching chemical accuracy can require hundreds of frames of force and energy labels. We propose a multi-stage computational framework -- ASTEROID, which lowers the data cost of MLFFs by leveraging a combination of cheap inaccurate data and expensive accurate data.
  arXiv  Detail & Related papers  (2023-06-05T04:34:54Z)
• Convolutional Monge Mapping Normalization for learning on sleep data [63.22081662149488]
  We propose a new method called Convolutional Monge Mapping Normalization (CMMN) CMMN consists in filtering the signals in order to adapt their power spectrum density (PSD) to a Wasserstein barycenter estimated on training data. Numerical experiments on sleep EEG data show that CMMN leads to significant and consistent performance gains independent from the neural network architecture.
  arXiv  Detail & Related papers  (2023-05-30T08:24:01Z)
• NeuralNEB -- Neural Networks can find Reaction Paths Fast [7.7365628406567675]
  Quantum mechanical methods like Density Functional Theory (DFT) are used with great success alongside efficient search algorithms for studying kinetics of reactive systems. Machine Learning (ML) models have turned out to be excellent emulators of small molecule DFT calculations and could possibly replace DFT in such tasks. In this paper we train state of the art equivariant Graph Neural Network (GNN)-based models on around 10.000 elementary reactions from the Transition1x dataset.
  arXiv  Detail & Related papers  (2022-07-20T15:29:45Z)
• BIGDML: Towards Exact Machine Learning Force Fields for Materials [55.944221055171276]
  Machine-learning force fields (MLFF) should be accurate, computationally and data efficient, and applicable to molecules, materials, and interfaces thereof. Here, we introduce the Bravais-Inspired Gradient-Domain Machine Learning approach and demonstrate its ability to construct reliable force fields using a training set with just 10-200 atoms.
  arXiv  Detail & Related papers  (2021-06-08T10:14:57Z)
• A Universal Framework for Featurization of Atomistic Systems [0.0]
  Reactive force fields based on physics or machine learning can be used to bridge the gap in time and length scales. We introduce the Gaussian multi-pole (GMP) featurization scheme that utilizes physically-relevant multi-pole expansions of the electron density around atoms. We demonstrate that GMP-based models can achieve chemical accuracy for the QM9 dataset, and their accuracy remains reasonable even when extrapolating to new elements.
  arXiv  Detail & Related papers  (2021-02-04T03:11:00Z)
• Targeted free energy estimation via learned mappings [66.20146549150475]
  Free energy perturbation (FEP) was proposed by Zwanzig more than six decades ago as a method to estimate free energy differences. FEP suffers from a severe limitation: the requirement of sufficient overlap between distributions. One strategy to mitigate this problem, called Targeted Free Energy Perturbation, uses a high-dimensional mapping in configuration space to increase overlap.
  arXiv  Detail & Related papers  (2020-02-12T11:10:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.