Materials Learning Algorithms (MALA): Scalable Machine Learning for Electronic Structure Calculations in Large-Scale Atomistic Simulations
This provides a tool for materials researchers to accelerate electronic structure calculations, enabling modeling of complex material systems beyond standard DFT limits, though it is incremental as it builds on existing local descriptor methods.
The authors tackled the computational expense of density functional theory (DFT) in large-scale atomistic simulations by developing the MALA package, a scalable machine learning framework that predicts electronic observables like local density of states and total energy, demonstrating its capabilities on systems such as boron clusters and a beryllium slab with scaling analyses.
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.