ICME Gap Analysis in Materials Informatics: Databases, Machine Learning, and Data-Driven Design: Session II
Sponsored by: TMS Materials Processing and Manufacturing Division, TMS: Computational Materials Science and Engineering Committee, TMS: Integrated Computational Materials Engineering Committee
Program Organizers: James Saal, Citrine Informatics; Carelyn Campbell, National Institute of Standards and Technology; Raymundo Arroyave, Texas A&M University
Wednesday 2:00 PM
February 26, 2020
Room: 30D
Location: San Diego Convention Ctr
Session Chair: Carelyn Campbell, National Institute for Standards and Technology; James Saal, Citrine Informatics
2:00 PM Invited
Gaps and Barriers to the Successful Integration and Adoption of Practical Materials Informatics Tools and Workflows: David McDowell1; 1Georgia Institute of Technology
Preparing the future workforce to contribute to Integrated Computational Materials Engineering (ICME) is most effectively pursued in the context of a materials innovation ecosystem that spans across conventional engineering, science, and computing disciplines. The ICME foundation spans principles of materials science with computational methods, including increasing reliance on modern data science methods that are savvy to digital information regarding hierarchical material structure and its relation to both process path and responses or properties. We consider gaps in the academic research enterprise regarding the connection of multiscale modeling of materials to decision support for materials development, and uncertainty quantification of both experiments and computation that drive ICME development processes. We consider barriers presented by limited duration research programs and lack of emphasis on underlying translational technologies across materials classes. We will discuss opportunities for materials innovation that can build the basis for practical materials informatics via novel instructional and curricular platforms.
2:40 PM Invited
Combining Machine Learning and ICME for Alloy Development: Bryce Meredig1; 1Citrine Informatics
Machine learning (ML) has a number of unique advantages that make it an effective complement to existing ICME tools. Specifically, ML is (1) computationally cheaper than physics-based simulations; (2) able to flexibly model (in principle) any input data and output properties; and (3) useful for performing uncertainty-aware optimization. In this talk, we will present a perspective on using ML to orchestrate multifidelity ICME simulations and experiments to accelerate alloy development.
3:20 PM
Magicmat (MAterials Genome and Integrated Computational MAterials Toolkit) and Its Application for Thermoelectric Materials Design: Changning Niu1; Ramya Gurunathan1; Abhinav Saboo1; Jiadong Gong1; 1QuesTek Innovations LLC
Managing and analyzing complex data form a cornerstone of materials engineering and design. We propose a toolkit for materials engineering and design which provides a workflow for achieving the core tasks in materials genome and ICME studies. A toolkit for easy-to-use data mining from complex databases with reduced user barriers would highly advance the goals of the MGI, while one that stores and manages a database with materials process-structure-property data, and assists modeling the linkages is critical for next-generation ICME activities. In this presentation, we will introduce the latest toolkit developed at QuesTek: MAterials Genome and Integrated Computational MAterials Toolkit, or magicmat. We will demonstrate its application for the design of thermoelectric materials.
3:40 PM Break
4:00 PM Invited
A Bayesian Framework for Materials Knowledge Systems: Surya Kalidindi1; 1Georgia Institute of Technology
This paper presents a new Bayesian framework that could guide the systematic application of the emerging toolsets of machine learning in the efforts to address two of the central bottlenecks encountered in materials innovation efforts: (i) the capture of core materials knowledge in reduced-order forms that allow one to rapidly explore the vast materials design spaces, and (ii) objective guidance in the selection of experiments or simulations needed to identify the governing physics in the materials phenomena of interest. The author’s perspective on gaps and barriers to the successful integration and adoption of the emergent materials analytics/informatics tools and workflows in material innovation efforts will be discussed.
4:40 PM Invited
Deep Materials Informatics: Illustrative Applications of Deep Learning in Materials Science: Ankit Agrawal1; 1Northwestern University
The growing application of data-driven analytics in materials science has led to the rise and popularity of the relatively new field of materials informatics. Within the arena of data analytics, in recent years deep learning has emerged as a game-changing technique, which has enabled numerous real-world applications such as self-driving cars. In this talk, I would present some of our recent works at the intersection of deep learning and materials informatics, for exploring processing-structure-property-performance (PSPP) linkages in materials. Illustrative examples include learning the chemistry of materials using only elemental composition, learning multiscale homogenization and localization linkages in high-contrast composites, and deep adversarial learning for microstructure design. The increasingly availability of materials databases and big data in general, along with groundbreaking advances in data science approaches offers lot of promise to accelerate the discovery, design, and deployment of next-generation materials.