machine learning

ACerS celebrates the International Year of Glass, plus more inside May 2022 ACerS Bulletin

By Lisa McDonald / April 21, 2022

The May 2022 issue of the ACerS Bulletin—featuring past, present, and future applications of glass—is now available online. Plus—National Day of Glass recap.

Read More

Determine oxidation stability of materials at MAX speed

By Lisa McDonald / April 23, 2021

Determining oxidation stability of new MAX phases is a difficult and expensive process with current computational and experimental methods. Researchers at Texas A&M University designed a new machine-learning-based scheme for predicting the oxidation of MAX phases at high temperatures, allowing them to conduct studies that may otherwise take years to perform.

Read More

Glass discovery and design: 21 challenges in artificial intelligence and machine learning for glass science

By Lisa McDonald / February 23, 2021

Artificial intelligence and machine learning approaches to materials design can accelerate the discovery of new glasses in an economical fashion. Researchers from the Indian Institute of Technology Delhi and the University of California, Los Angeles, identified 21 challenges that, when addressed, can aid in harnessing the full potential of these methods for glass science.

Read More

Machine learning model predicts superhard materials from crystal structure

By Lisa McDonald / September 25, 2020

Conventionally, theoretical models are unable to predict a material’s hardness from its crystal structure because the underlying physical principles are complex. A new machine learning model developed by two researchers at Skolkovo Institute of Science and Technology succeeds in making such predictions in a fast and reliable manner.

Read More

A double-edged sword—reverse engineered 3D-printed parts show security risk presented by machine learning

By Lisa McDonald / July 31, 2020

Machine learning is poised to play a big role in speeding up materials discovery and commercialization—but could such techniques present a risk to the global additive manufacturing market as well? Researchers at New York University showed they could potentially steal trade secrets by reverse engineering 3D-printed parts using machine learning.

Read More

Predicting optimal glass compositions: A review of machine learning for glass science and engineering

By Lisa McDonald / September 17, 2019

Machine learning can greatly facilitate design of new glasses by predicting a range of promising compositions to test. A recent paper by researchers from the University of California, Los Angeles, reviews studies investigating machine learning methods for just that purpose.

Read More

Other materials stories that may be of interest

By Lisa McDonald / February 13, 2019

Nanoparticles may promote cancer metastasis, ceramic sensors for bridge strain, and other materials stories that may be of interest for February 13, 2019.

Read More

Other materials stories that may be of interest

By Faye Oney / October 24, 2018

Fast-charging stations for electric vehicles, aluminum air flow batteries, and other materials stories that may be of interest for October 24, 2018.

Read More

Disorderly conduct: Insight into materials failure could lead to glasses that are less likely to break

By April Gocha / December 5, 2017

An interdisciplinary group of scientists at the University of Pennsylvania have harnessed intense computation, data, and modeling power to determine how disordered solids fail, an understanding that could help engineer custom materials, such as glass that is less likely to break.

Read More

Other materials stories that may be of interest

By April Gocha / March 1, 2017

Atom-scale oxidation mechanism of nanoparticles helps develop anti-corrosion materials, unique workflow to design new materials, and other materials stories that may be of interest for March 1, 2017.

Read More