AI learned to select materials for research

The new algorithm will help researchers speed up their experiments. An AI-based model can tell in a split second what materials are suitable for a scientist for a project.

The researchers explained that they are always on the lookout for materials with very specific properties for their projects. But usually, there are too many options and they cannot constantly experiment to figure out which material is right for them. Moreover, it stretches the duration of the research several times.

Therefore, scientists have created an algorithm using artificial intelligence (AI) that automatically finds material for any project. A team of researchers from Carnegie Mellon University and the University of Calgary has refined the algorithm, allowing them to quickly and accurately find materials with the desired properties.

“Since the number of materials is so large, it is very difficult to experimentally and computationally study the properties of materials. Therefore, we have created algorithms that can quickly predict their properties. ”

researchers from Carnegie Mellon University
Since the algorithm can predict the properties of a wide range of materials, it can be applied beyond research. For example, an algorithm can find a material with thermal properties suitable for solar panels. In addition, he can identify the materials that are needed to make drugs. To use the algorithm, a researcher can simply enter a query and find the property he is looking for.

“You can use this algorithm and train your deep learning model so that it will show you what you want in a few seconds. In this way, it will predict the properties of various materials with high accuracy and will be applied in many areas. “