A Data Ecosystem to Support Machine Learning in Materials Science

TL;DR

This paper introduces the Materials Data Facility and Data and Learning Hub for Science, two platforms that enhance data sharing and integration for machine learning applications in materials science.

Contribution

It presents two new data ecosystems, MDF and DLHub, that facilitate data discovery, dissemination, and linking with machine learning models in materials science.

Findings

MDF and DLHub enable efficient data access and sharing.

They support linking data with machine learning models.

Users can access capabilities via web and APIs.

Abstract

Facilitating the application of machine learning to materials science problems will require enhancing the data ecosystem to enable discovery and collection of data from many sources, automated dissemination of new data across the ecosystem, and the connecting of data with materials-specific machine learning models. Here, we present two projects, the Materials Data Facility (MDF) and the Data and Learning Hub for Science (DLHub), that address these needs. We use examples to show how MDF and DLHub capabilities can be leveraged to link data with machine learning models and how users can access those capabilities through web and programmatic interfaces.