用机器学习加速大尺度材料模拟中的电子结构计算
Materials Learning Algorithms (MALA): Scalable Machine Learning for Electronic Structure Calculations in Large-Scale Atomistic Simulations
- 基于原子局部环境描述符构建模型，预测电子态密度等关键量
- 支持从数据采样到推理的全流程，可处理大规模原子体系
- 适合复杂材料系统研究，尤其适合传统DFT难以企及的场景
我们提出Materials Learning Algorithms (MALA)框架，一个用于大规模原子模拟中加速密度泛函理论（DFT）计算的可扩展机器学习系统。该框架利用原子局部环境的描述符，高效预测局域态密度、电子密度、总能等关键电子观测量。MALA将数据采样、模型训练与可扩展推断集成于统一库中，并兼容标准DFT和分子动力学软件。通过硼簇、铝在固液相变边界、以及大型铍晶格堆垛层错的电子结构预测等案例，展示了其能力。缩放分析揭示了其计算效率并指出了未来优化瓶颈。凭借在远超传统DFT规模下建模电子结构的能力，MALA成为先进材料研究的通用工具。
原文摘要 · Abstract (English)
We present the Materials Learning Algorithms (MALA) package, a scalable machine learning framework designed to accelerate density functional theory (DFT) calculations suitable for large-scale atomistic simulations. Using local descriptors of the atomic environment, MALA models efficiently predict key electronic observables, including local density of states, electronic density, density of states, and total energy. The package integrates data sampling, model training and scalable inference into a unified library, while ensuring compatibility with standard DFT and molecular dynamics codes. We demonstrate MALA's capabilities with examples including boron clusters, aluminum across its solid-liquid phase boundary, and predicting the electronic structure of a stacking fault in a large beryllium slab. Scaling analyses reveal MALA's computational efficiency and identify bottlenecks for future optimization. With its ability to model electronic structures at scales far beyond standard DFT, MALA is well suited for modeling complex material systems, making it a versatile tool for advanced materials research.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。