AlphaNet: Scaling Up Local Frame-based Atomistic Foundation Model

2025-01-13

AI Summary

To address the trade-off among accuracy, efficiency, and scalability in machine learning force fields (MLFFs), this work proposes an SE(3)-equivariant atomic force field based on learnable local coordinate frames. Methodologically, it introduces a learnable local frame transformation to efficiently encode atomic geometric environments, integrated with multi-scale environmental representations and a strictly SE(3)-equivariant neural network architecture—ensuring both physical consistency and computational efficiency. Compared to NequIP and DeepPot, our model achieves significantly improved energy and force prediction accuracy across diverse benchmark tasks—including defective graphene, formic acid decomposition, zeolites, and surface reactions—while attaining superior accuracy-efficiency trade-offs and enhanced cross-scale generalization. The framework thus establishes a new paradigm for large-scale molecular dynamics simulations: highly accurate, strictly equivariant, and strongly scalable.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

We present AlphaNet, a local frame-based equivariant model designed to achieve both accurate and efficient simulations for atomistic systems. Recently, machine learning force fields (MLFFs) have gained prominence in molecular dynamics simulations due to their advantageous efficiency-accuracy balance compared to classical force fields and quantum mechanical calculations, alongside their transferability across various systems. Despite the advancements in improving model accuracy, the efficiency and scalability of MLFFs remain significant obstacles in practical applications. AlphaNet enhances computational efficiency and accuracy by leveraging the local geometric structures of atomic environments through the construction of equivariant local frames and learnable frame transitions. We substantiate the efficacy of AlphaNet across diverse datasets, including defected graphene, formate decomposition, zeolites, and surface reactions. AlphaNet consistently surpasses well-established models, such as NequIP and DeepPot, in terms of both energy and force prediction accuracy. Notably, AlphaNet offers one of the best trade-offs between computational efficiency and accuracy among existing models. Moreover, AlphaNet exhibits scalability across a broad spectrum of system and dataset sizes, affirming its versatility.

Problem

• Machine Learning Force Fields
• Accuracy Improvement
• Large System Scalability

Innovation

• AlphaNet Model
• Atomic Environment Understanding
• Accuracy and Speed Improvement