Large-scale deep learning models accelerate the discovery of stable new crystals

Large-scale deep learning models accelerate the discovery of stable new crystals
Image: Intellect (generated)

In short: Researchers using large-scale graph networks have significantly sped up the search for new inorganic crystals. By training models on vast amounts of data, the team identified 2.2 million stable structures, many of which were previously unknown to human intuition, representing a massive expansion in our known library of stable materials.

The study utilized graph networks trained on large-scale data, including 48,000 known stable crystals and hundreds of millions of first-principles calculations. This approach allowed the researchers to predict millions of new stable structures. While the model demonstrated high accuracy in predicting ionic conductivity and interatomic potentials, the findings rely on the quality of the underlying large-scale computational data used for training.

Why it matters: AI-driven large-scale modeling can expand our known universe of stable materials by orders of magnitude.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/technology-research-large-scale-deep-learning-models-accelerate-the-discovery-of.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…