Skip to main content
U.S. flag

An official website of the United States government

Dot gov

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Https

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Machine Learning: Educating the Next Generation Materials Workforce

Summary

We are educating the next generation workforce with MGI skills and providing tools to rapidly deploy these skills. This includes knowledge and tools for machine learning as well as integration of data, computation, and in-lab experiments to accelerate materials optimization and discovery. Our customers include scientists and engineers from industry, academia, and national labs. Two main projects include the annual Machine Learning and Materials Research Bootcamp and the Resource for Materials Informatics community website.

Description

Annual Bootcamp: Machine Learning for Materials Research (MLMR)

A photo of attendees attending MLMR Bootcamp 2018.

The fifth annual MLMR met in the summer of 2020 with 180 attendees from 12 countries, 30% of whom were from industry. Over the 5 years of the bootcamp, we have had attendees from a total of 19 countries. We also run tutorials at MRS, APS, MLSE, NSF meetings, among others.

Bootcamp description

Four days of lectures and hands-on exercises covering a range of data analysis topics from introduction to python and data pre-processing to advanced machine learning analysis techniques. Example topics include:

  • Identifying important features in complex/high dimensional data
  • Visualizing high dimensional data to facilitate user analysis.
  • Identifying the 'descriptors' that best predict variance in functional properties.
  • Quantifying similarities between materials using complex/high dimensional data
  • Identifying the most informative experiment to perform next.

Hands-on exercises

Hands-on exercises include practical use of machine learning tools on real materials experimental data (scalar values, spectra, micrographs, etc.) On day 4, scientists demonstrate how they performed recently published research, from loading and preprocessing data to analyzing and visualizing results, all in Jupyter notebooks.

REMI: REsource for Materials Informatics

REMI: REsource for Materials Informatics

REMI is a repository for tutorials and code examples covering materials data import/export, pre-processing, and analysis. Search by material system, synthesis / simulation method, measurement method, data type, data analysis type, and more. With this site we seek to create an open community for sharing knowledge and tools.

Created August 31, 2020, Updated September 4, 2020