Advancements in Machine Learning Revolutionize Material Modeling and Biology Research

Researchers from the Center for Advanced Systems Understanding (CASUS) and Sandia National Laboratories have developed a machine learning-based simulation method called Materials Learning Algorithms (MALA) that surpasses traditional electronic structure simulation techniques. MALA integrates machine learning with physics-based approaches to accurately predict the electronic structure of materials, enabling access to previously unattainable length scales. The software stack achieved a speedup of over 1,000 times for smaller system sizes and accurately performed electronic structure calculations involving more than 100,000 atoms. This breakthrough opens up computational possibilities for addressing societal challenges and advancing applied research in areas such as drug design, energy storage, and semiconductor devices.