Graph Networks for Materials Exploration

In a recent research study published in Nature, authors Amil Merchant and Ekin Dogus Cubuk introduce GNoME, a deep learning tool that has discovered 2.2 million new crystals, including 380,000 stable materials. These stable materials have the potential to power future technologies such as superconductors, supercomputers, and next-generation batteries. GNoME has significantly increased the number of technologically viable materials known to humanity, and its predictions have been independently verified by external researchers. The authors have made GNoME’s predictions available to the research community and hope that it will drive further research into inorganic crystals and the potential of machine learning in materials discovery. GNoME’s discovery of 2.2 million new materials is an unprecedented achievement and demonstrates the power of AI in predicting materials at an unprecedented scale and level of accuracy.

https://deepmind.google/discover/blog/millions-of-new-materials-discovered-with-deep-learning/

To top