Our Sponsors

You are not logged in.

Machine Learning Models vs Simulation Models

Keynote Speaker (by invitation only) at Digital Twin Symposium 2019, presented by Amanda S Barnard


A fundamental aim of materials research is to identify features of materials that can be tuned to control how the material performs under specific application conditions. The combination of computational materials science with machine learning provides a powerful way of relating structural features with functional properties, but combining these fundamentally different scientific approaches is not as straightforward as it seems.  Machine learning methods were developed for large data sets with small numbers of consistent features.  Typical materials data sets are small, with high dimensionality and high variance in the feature space, and suffer from numerous destructive biases.  None of the established data science or machine learning methods in widespread use today were devised with materials data sets in mind, but there are ways to overcome these issues and use them reliably.  In this presentation we will discuss the impact of domain-specific constraints on data-driven materials design, and explore the differences between materials simulation and materials informatics that can be leveraged for greater impact. We will review the differences between machine learning models and simulations models, and discuss feature engineering, dimension reduction, prediction and visualisation.  We will conclude by discussing surrogate models and types of model-informed machine learning, and their potential role in a digital twin.

Presenting Author

Photo ofAmanda S Barnard

Amanda S Barnard


Dr Amanda Barnard is a Chief Research Scientist in CSIRO's Data61. She received her Ph.D. in Physics in 2003, followed by a Distinguished Postdoctoral Fellow in the Center for Nanoscale Materials at ANL, and a senior Violette & Samuel Glasstone Fellow at the University of Oxford. She joined CSIRO as an ARC Queen Elizabeth II Fellow in 2009, and then as an OCE Science Leader, where she lead research developing structure/property relationships using computational modeling, machine learning and AI. Dr Barnard is on the Editorial Advisory Board for Nanoscale, the Senior Editorial Board for the Journal of Physics: Materials, the International Executive Board of Nano Futures, the Board of Directors for New Zealand eScience Infrastructure (NeSI), and the Chair of the National Computational Merit Allocation Scheme for Australia, a member of Australia’s National Computational Infrastructure (NCI) capital refresh Procurement Steering Committee and the Chair of the User Reference Group for the Pawsey Supercomputing Centre capital refresh. She also sits on the advisory boards of the Centre for Biomedical Data Visualisation (BioViS) at the Garvan Institute, Centre for Theoretical and Computational Molecular Science (CTCMS) at the University of Queensland, and ChoiceFlows Inc. For her work she has won 12 national and international awards, including the 2009 Young Scientist Prize in Computational Physics from IUPAP, the 2009 Australian Physical Scientist of the Year, the 2010 Frederick White Prize from the Australian Academy of Sciences, the 2014 ACS Nano Lectureship (Asia/Pacific) from the American Chemical Society, and the 2014 Feynman Prize in Nanotechnology (Theory) from the Foresight Institute, being the first woman to do so in the history of the award.