deep learning

Freely available deep learning method ‘fills in the blank’ of unknown internal material structures

By Lisa McDonald / May 9, 2023

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

Finding trees within the forest: Deep learning network detects individual carbon nanotubes in SEM images

By Lisa McDonald / March 21, 2023

For researchers to improve the properties of carbon nanotubes grown en masse, they must first be able to measure and characterize how individual nanotubes are assembled within carbon nanotube “forests.” In a recent paper, researchers at the University of Missouri outlined a deep learning technique to segment these forests in scanning electron microscopy images.

Read More

Deep learning provides deep help—researchers develop publicly available software for rational design of oxide glasses

By Lisa McDonald / May 5, 2020

Designing new oxide glass compositions can be an arduous process when relying on the “cook and look” approach. Researchers at the Indian Institute of Technology Delhi developed composition-property deep learning models for eight key oxide glass properties, and they made the software available publicly online.

Read More

ICACC 2018—Never a dull moment between sessions, networking, and government shutdown

By Eileen De Guire / January 26, 2018

ICACC organizers worked around havoc created by the federal shutdown to deliver a conference with excellent technical content, networking, and an expo.

Read More