Mixture of Experts Framework in Machine Learning Interatomic Potentials for Atomistic Simulations

• URL: http://arxiv.org/abs/2604.26143v1
• Date: Tue, 28 Apr 2026 22:04:05 GMT
• Title: Mixture of Experts Framework in Machine Learning Interatomic Potentials for Atomistic Simulations
• Abstract summary: First-principles atomistic simulations are essential for understanding complex material phenomena but are fundamentally limited by their computational cost.<n>We introduce a multifidelity "Mixture-of-Experts" framework based on the E(3)-equivariant Allegro architecture.
• Score: 0.3024996106292169
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: First-principles atomistic simulations are essential for understanding complex material phenomena but are fundamentally limited by their computational cost. While Machine Learning Interatomic Potentials (MLIPs) have drastically improved cost for a given accuracy, their inference cost remains a bottleneck for massive systems or long timescales. To address this, we introduce a multifidelity "Mixture-of-Experts" framework based on the E(3)-equivariant Allegro architecture. Our method spatially partitions the simulation domain into a chemically complex region (e.g., reactive interfaces) and a simple region (e.g., bulk lattice), assigning models of varying capacity to each. Among the challenges in such static domain decomposition, the mechanical mismatch between models at the interface is particularly critical, as it can generate artificial stress fields and instability. We address this challenge with a co-training strategy in which the loss function includes agreement constraints -- penalties on per-atom energy and force discrepancies between models evaluated on shared bulk environments -- forcing the independent models to learn a consistent physical description of the bulk material. We validate this approach on a realistic Pt+CO catalytic system, demonstrating that the co-trained models maintain exact energy conservation, align their bulk mechanical response (e.g., equation of state and bulk modulus), and achieve predictive accuracy comparable to a full high-fidelity simulation at more than twice the computational speed.

Related papers

• A Physics-Grounded Benchmark for Multi-Agent Dynamics in World Models [96.64934195520802]
  We introduce CrashTwin, a physics-grounded evaluation framework designed to stress-test the physical trustworthiness of world models.<n>CrashTwin couples a diverse dataset of multi-agent collision scenarios, comprising 25K controllable synthetic 12K in-the-wild real-world collision sequences.<n>We propose a diagnostic suite that systematically evaluates three dimensions:-temporal consistency, momentum and kinetic energy conservation, and world-dynamics integrity.
  arXiv  Detail & Related papers  (2026-06-27T06:13:35Z)
• A Hybrid GNN-FEM Framework for Phase-Field Fracture Simulation. Physics-Preserving Hybridization for Generalizable Surrogate Modeling [1.1470070927586018]
  We propose a hybrid GNN--FEM framework for efficient and generalizable phase-field fracture modeling.<n>The proposed framework achieves strong generalization across varying geometries, loading conditions, material properties, and discretizations.<n> Numerical experiments demonstrate that the hybrid approach reduces computational cost while maintaining accuracy compared with conventional FEM.
  arXiv  Detail & Related papers  (2026-06-12T10:02:28Z)
• Advancing Practical Quantum Embedding Simulations via Operator Commutativity Based State Preparation for Complex Chemical Systems [0.0]
  Hybrid quantum-classical algorithms implemented on quantum computers offer a promising route toward addressing this problem.<n>We propose a dynamic ansatz construction strategy based on operator commutativity and energy driven screening within density matrix embedding theory.<n>We demonstrate improved accuracy and significantly reduced quantum gate requirements compared with conventional ansatze.
  arXiv  Detail & Related papers  (2026-04-21T13:51:55Z)
• From Complex Dynamics to DynFormer: Rethinking Transformers for PDEs [6.873342825786888]
  Transformer-based neural operators have emerged as powerful data-driven alternatives.<n>We propose DynFormer, a novel dynamics-informed neural operator.<n>We show that DynFormer achieves up to a 95% reduction in relative error compared to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-03-03T15:45:09Z)
• Equivariant Evidential Deep Learning for Interatomic Potentials [55.6997213490859]
  Uncertainty quantification is critical for assessing the reliability of machine learning interatomic potentials in molecular dynamics simulations.<n>Existing UQ approaches for MLIPs are often limited by high computational cost or suboptimal performance.<n>We propose textitEquivariant Evidential Deep Learning for Interatomic Potentials ($texte2$IP), a backbone-agnostic framework that models atomic forces and their uncertainty jointly.
  arXiv  Detail & Related papers  (2026-02-11T02:00:25Z)
• PhyG-MoE: A Physics-Guided Mixture-of-Experts Framework for Energy-Efficient GNSS Interference Recognition [49.955269674859004]
  This paper introduces PhyG-MoE (Physics-Guided Mixture-of-Experts), a framework designed to align model capacity with signal complexity.<n>Unlike static architectures, the proposed system employs a spectrum-based gating mechanism that routes signals based on their spectral feature entanglement.<n>A high-capacity TransNeXt expert is activated on-demand to disentangle complex features in saturated scenarios, while lightweight experts handle fundamental signals to minimize latency.
  arXiv  Detail & Related papers  (2026-01-19T07:57:52Z)
• Efficient Exploration of Chemical Kinetics [0.0]
  Estimating reaction rates and chemical stability is fundamental, yet efficient methods for large-scale simulations remain out of reach.<n>This dissertation presents a holistic solution, co-designing physical representations, statistical models, and systems architecture in the Optimal Transport Gaussian Process framework.
  arXiv  Detail & Related papers  (2025-10-24T11:52:08Z)
• A Neural Operator based Hybrid Microscale Model for Multiscale Simulation of Rate-Dependent Materials [0.7946947383637114]
  We employ neural operators to predict the microscale physics, resulting in a hybrid model that combines data-driven and physics-based approaches.<n>We apply this method to time-dependent solid mechanics problems involving viscoelastic state is represented by internal variables only at the microscale.
  arXiv  Detail & Related papers  (2025-06-20T11:25:26Z)
• Coupled reaction and diffusion governing interface evolution in solid-state batteries [4.707991478885645]
  We conduct large-scale explicit reactive simulations with quantum accuracy for a symmetric battery cell.<n>We explain experimental observations of the SEI formations and elucidate the Li creep mechanisms, critical to dendrite initiation.<n>Our approach is to crease a digital twin from first principles, without adjustable parameters fitted to experiment.
  arXiv  Detail & Related papers  (2025-06-12T17:49:05Z)
• GausSim: Foreseeing Reality by Gaussian Simulator for Elastic Objects [55.02281855589641]
  GausSim is a novel neural network-based simulator designed to capture the dynamic behaviors of real-world elastic objects represented through Gaussian kernels.<n>We leverage continuum mechanics and treat each kernel as a Center of Mass System (CMS) that represents continuous piece of matter.<n>In addition, GausSim incorporates explicit physics constraints, such as mass and momentum conservation, ensuring interpretable results and robust, physically plausible simulations.
  arXiv  Detail & Related papers  (2024-12-23T18:58:17Z)
• Multi-fidelity Hierarchical Neural Processes [79.0284780825048]
  Multi-fidelity surrogate modeling reduces the computational cost by fusing different simulation outputs. We propose Multi-fidelity Hierarchical Neural Processes (MF-HNP), a unified neural latent variable model for multi-fidelity surrogate modeling. We evaluate MF-HNP on epidemiology and climate modeling tasks, achieving competitive performance in terms of accuracy and uncertainty estimation.
  arXiv  Detail & Related papers  (2022-06-10T04:54:13Z)
• Interfacing Finite Elements with Deep Neural Operators for Fast Multiscale Modeling of Mechanics Problems [4.280301926296439]
  In this work, we explore the idea of multiscale modeling with machine learning and employ DeepONet, a neural operator, as an efficient surrogate of the expensive solver. DeepONet is trained offline using data acquired from the fine solver for learning the underlying and possibly unknown fine-scale dynamics. We present various benchmarks to assess accuracy and speedup, and in particular we develop a coupling algorithm for a time-dependent problem.
  arXiv  Detail & Related papers  (2022-02-25T20:46:08Z)
• Hybridized Methods for Quantum Simulation in the Interaction Picture [69.02115180674885]
  We provide a framework that allows different simulation methods to be hybridized and thereby improve performance for interaction picture simulations. Physical applications of these hybridized methods yield a gate complexity scaling as $log2 Lambda$ in the electric cutoff. For the general problem of Hamiltonian simulation subject to dynamical constraints, these methods yield a query complexity independent of the penalty parameter $lambda$ used to impose an energy cost.
  arXiv  Detail & Related papers  (2021-09-07T20:01:22Z)

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.