Machine Learning for Materials Discovery: A New and Promising Approach

Scientists use machine-learning to accelerate material discovery

The new approach will help to improve the understanding of different carbon states and lead the search for yet-to-be discovered materials.

We use, wear, eat, and create materials. We sometimes invent materials by accident, as with Silly Putty. Making useful materials can be a costly and time-consuming process of trial and errors.

Scientists from the U.S. Department of Energy (DOE) Argonne Laboratory recently demonstrated a process that combines machine learning (ML), a form of artificial intelligence, and high-performance computing to identify and explore promising new materials. This new approach may help to accelerate the design and discovery of useful materials.