Toward Interpretable Machine Learning Models for Materials Discovery

zu Verbundenen Objekten

Machine learning (ML) and artificial intelligence (AI) methods for modeling useful materials properties are now important technologies for rational design and optimization of bespoke functional materials. Although these methods make good predictions of the properties of new materials, current modeling methods use efficient but rather arcane (difficult‐to‐interpret) mathematical features (descriptors) to characterize materials. Data‐driven ML models are considerably more useful if more chemically interpretable descriptors are used to train them, as long as these models also accurately recapitulate the properties of materials in training and test sets used to generate and validate the models. Herein, how a particular type of molecular fragment descriptor, the signature descriptor, achieves these joint aims of accuracy and interpretability is described. Seven different types of materials properties are modeled, and the performance of models generated from signature descriptors is compared with those generated by widely used Dragon descriptors. The key descriptors in the model represent functionalities that make chemical sense. Mapping these fragments back on to exemplar materials provides a useful guide to chemists wishing to modify promising lead materials to improve their properties. This is one of the first applications of signature descriptors to the modeling of complex materials properties.

Standort
Deutsche Nationalbibliothek Frankfurt am Main
Umfang
Online-Ressource
Sprache
Englisch

Erschienen in
Toward Interpretable Machine Learning Models for Materials Discovery ; volume:1 ; number:8 ; year:2019 ; extent:16
Advanced intelligent systems ; 1, Heft 8 (2019) (gesamt 16)

Urheber
Mikulskis, Paulius
Alexander, Morgan R.
Winkler, David Alan

DOI
10.1002/aisy.201900045
URN
urn:nbn:de:101:1-2022071507553776634418
Rechteinformation
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
Letzte Aktualisierung
15.08.2025, 07:31 MESZ

Datenpartner

Dieses Objekt wird bereitgestellt von:
Deutsche Nationalbibliothek. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.
Original beim Datenpartner anzeigen

Beteiligte

  • Mikulskis, Paulius
  • Alexander, Morgan R.
  • Winkler, David Alan

Ähnliche Objekte (12)

Model-agnostic Interpretable Machine Learning

zweidimensionales bewegtes Bild

Model-agnostic Interpretable Machine Learning

Interpretable machine learning for genomics

Interpretable machine learning for genomics

Model-agnostic interpretable machine learning

Model-agnostic interpretable machine learning

R.ROSETTA: an interpretable machine learning framework

R.ROSETTA: an interpretable machine learning framework

Ultra-fast interpretable machine-learning potentials

Ultra-fast interpretable machine-learning potentials

Conceptual challenges for interpretable machine learning

Conceptual challenges for interpretable machine learning

Ultra-fast interpretable machine-learning potentials

Ultra-fast interpretable machine-learning potentials

Discovery of Unconventional Proton‐Conducting Inorganic Solids via Defect‐Chemistry‐Trained, Interpretable Machine Learning

Discovery of Unconventional Proton‐Conducting Inorganic Solids via Defect‐Chemistry‐Trained, Interpretable Machine Learning

Co-training machine learning enables interpretable discovery of near-infrared phosphors with high performance

Co-training machine learning enables interpretable discovery of near-infrared phosphors with high performance

Leveraging interpretable machine learning in intensive care

Leveraging interpretable machine learning in intensive care

Interpretable Machine Learning Approaches in Computational Biology

Interpretable Machine Learning Approaches in Computational Biology

Interpretable Machine Learning of Two‐Photon Absorption

Interpretable Machine Learning of Two‐Photon Absorption

Verbundene Objekte