What if you could predict a material’s internal microstructure based solely on its external surface characteristics? A new deep learning method developed at Massachusetts Institute of Technology provides such a capability, and all data and codes used for the study are freely available for anyone to use through GitHub.
Read More
Current neural networks are incapable of understanding symmetry, which limits the conclusions that can be drawn from the data. Researchers from Lehigh and Stanford universities developed a new model that includes symmetry-aware features to improve symmetry approximation.
Read More
A new sponge-like material structure developed by scientists at MIT can soak up the sun—and uses that sunlight to make steam.
Read More