The new method can determine crystal structures underlying experimental data thus far difficult to analyze. A joint research team led by Yuuki Kubo and Shiji Tsuneyuki of the University of Tokyo has ...

A research team from the Institute of Statistical Mathematics and Panasonic Holdings Corporation has developed a machine learning algorithm, ShotgunCSP, that enables fast and accurate prediction of ...

Perchlorate compounds are known for their explosive nature. To understand what makes these compounds so explosive, a team of researchers developed a novel deep learning-based method that analyses ...

Duplicates of crystal structures are flooding databases, implicating repositories hosting organic, inorganic, and computer-generated crystals. The issue raises questions about curation practices at ...

The proposed novel method uses deep learning to study the physical properties of compounds such as explosive perchlorates by using only their crystal structure and thus avoiding dangerous experiments.