Global Universal Scaling and Ultra-Small Parameterization in Machine Learning Interatomic Potentials with Super-Linearity

2025-02-11

AI Summary

Current machine learning interatomic potentials (MLIPs) lack physical constraints, resulting in poor out-of-distribution generalization, parameter redundancy, and limited transferability across chemical compositions and crystal structures. To address these limitations, we propose SUS2-MLIP—a novel MLIP framework that synergistically integrates a universal equation of state (UEOS)-driven global scaling law with a superlinear embedding interaction function. This design enables decoupled parametrization in elemental and coordinate spaces, coupled with nonlinear feature mapping. Consequently, SUS2-MLIP achieves intrinsic physical scalability, drastically reduces parameter count, and maintains strong generalization even under data-scarce conditions. On multi-component materials, SUS2-MLIP consistently outperforms state-of-the-art MLIPs in both accuracy and robustness: it delivers high-fidelity predictions for energy and atomic forces while ensuring physical consistency across compositions and structures. Moreover, it achieves several-fold speedup in computational efficiency without sacrificing predictive fidelity.

Technology Category

• Machine Learning: Scalability of ML Systems
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Natural Language Processing: Safety and Robustness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations

Abstract

Using machine learning (ML) to construct interatomic interactions and thus potential energy surface (PES) has become a common strategy for materials design and simulations. However, those current models of machine learning interatomic potential (MLIP) provide no relevant physical constrains, and thus may owe intrinsic out-of-domain difficulty which underlies the challenges of model generalizability and physical scalability. Here, by incorporating physics-informed Universal-Scaling law and nonlinearity-embedded interaction function, we develop a Super-linear MLIP with both Ultra-Small parameterization and greatly expanded expressive capability, named SUS2-MLIP. Due to the global scaling rooting in universal equation of state (UEOS), SUS2-MLIP not only has significantly-reduced parameters by decoupling the element space from coordinate space, but also naturally outcomes the out-of-domain difficulty and endows the potentials with inherent generalizability and scalability even with relatively small training dataset. The nonlinearity-enbeding transformation for interaction function expands the expressive capability and make the potentials super-linear. The SUS2-MLIP outperforms the state-of-the-art MLIP models with its exceptional computational efficiency especially for multiple-element materials and physical scalability in property prediction. This work not only presents a highly-efficient universal MLIP model but also sheds light on incorporating physical constraints into artificial-intelligence-aided materials simulation.

Problem

• Enhance ML interatomic potential generalizability
• Reduce parameters in MLIP models
• Improve computational efficiency in materials simulation

Innovation

• Physics-informed Universal-Scaling law
• Nonlinearity-embedded interaction function
• Ultra-Small parameterization technique