XenonPy.MDL—Comprehensive Library of Pre-trained Models for Predicting Materials Properties

—Transfer Learning May Serve as a Solution to the Scarcity of Materials Property Data—

2019.10.01


Research Organization of Information and Systems
National Institute for Materials Science (NIMS)

A joint research group consisting of the Institute of Statistical Mathematics (ISM) and the National Institute for Materials Science (NIMS) has developed approximately 140,000 machine learning models capable of predicting 45 different types of physical properties in small molecules, polymers and inorganic materials. The joint group then made XenonPy.MDL—a pre-trained model library—publicly available.

(”Predicting materials properties with little data using shotgun transfer learning” Hironao Yamada, Chang Liu, Stephen Wu, Yukinori Koyama, Shenghong Ju, Junichiro Shiomi, Junko Morikawa, Ryo Yoshida; Journal: ACS Central Science [September 30, 2019]; DOI:10.1021/acscentsci.9b00804)

Abstract

A joint research group consisting of the Institute of Statistical Mathematics (ISM) and the National Institute for Materials Science (NIMS) has developed approximately 140,000 machine learning models capable of predicting 45 different types of physical properties in small molecules, polymers and inorganic materials. The joint group then made XenonPy.MDL—a pre-trained model library— publicly available.

XenonPy—an open source platform for materials informatics (MI) research—was jointly developed by NIMS and a team at the ISM Data Science Center for Creative Design and Manufacturing (Chang Liu (Project Assistant Professor), Yoh Noguchi (Project Researcher), Stephen Wu (Assistant Professor), Hironao Yamada (Project Researcher) and Ryo Yoshida (Center Director).

XenonPy uses machine learning algorithms to perform various tasks of MI. Users of XenonPy can run the pre-trained models available in the XenonPy.MDL library via the application programming interface (API) and use them to construct a variety of materials design workflows. The joint group recently reported the release of XenonPy.MDL in a research article published in ACS Central Science, a journal of the American Chemical Society.

In addition, as described in the article, the group succeeded in demonstrating the great potential of transfer learning to overcome the problem of limited amounts of materials data in various tasks of MI (i.e., predicting the physical properties of small molecules, polymers and inorganic crystalline materials using exceedingly limited materials data).

This research project was supported in part by the JST “Materials Research by Information Integration” Initiative (MI2I) and was conducted at NIMS between FY2015 and FY2019.


"figure: Thermophysical properties (i.e., thermal conductivity) of polymers predicted by transfer learning. The joint research group succeeded in constructing a machine learning model capable of the extrapolative prediction of three new polymers that resided in far tails of the training data distribution." Image

figure: Thermophysical properties (i.e., thermal conductivity) of polymers predicted by transfer learning. The joint research group succeeded in constructing a machine learning model capable of the extrapolative prediction of three new polymers that resided in far tails of the training data distribution.



Contacts

(Regarding this research)

Ryo Yoshida
Professor
Director of the Data Science Center for Creative Design and Manufacturing
The Institute of Statistical Mathematics
Research Organization of Information and Systems
Group Leader,
Materials Exploration Platform Group,
Center for Materials Research by Information Integration,
Research and Services Division of Materials Data and Integrated System (MaDIS),
National Institute for Materials Science (NIMS)
Tel: +81-50-5533-8534
E-Mail: yoshidar=ism.ac.jp
(Please change "=" to "@")

(General information)

URA Station
Planning Unit
Administration Planning and Coordination Section
Institute of Statistical Mathematics
Research Organization of Information and Systems
Tel: +81-50-5533-8580
Fax: +81-41-526-4348
E-Mail: ask-ura=ism.ac.jp
(Please change "=" to "@")
Public Relations Office
National Institute for Materials Sciences
Tel: +81-29-859-2026
Fax: +81-29-859-2017
E-Mail: pressrelease=ml.nims.go.jp
(Please change "=" to "@")