Accelerating materials property predictions using machine learning

Abstract

The materials discovery process can be significantly expedited and simplified if we can learn effectively from available knowledge and data. In the present contribution, we show that efficient and accurate prediction of a diverse set of properties of material systems is possible by employing machine (or statistical) learning methods trained on quantum mechanical computations in combination with the notions of chemical similarity. Using a family of one-dimensional chain systems, we present a general formalism that allows us to discover decision rules that establish a mapping between easily accessible attributes of a system and its properties. It is shown that fingerprints based on either chemo-structural (compositional and configurational information) or the electronic charge density distribution can be used to make ultra-fast, yet accurate, property predictions. Harnessing such learning paradigms extends recent efforts to systematically explore and mine vast chemical spaces, and can significantly accelerate the discovery of new application-specific materials.

Fulfilling the promise of the materials genome initiative with high-throughput experimental methodologies

“…The benefits of machine learning for accelerated materials data analysis have already been realized, with numerous studies showing the great potential for research and discovery. [199][200][201] These studies include a wide range of materials analysis challenges including crystal structure [202][203][204] and phase diagram 130,[205][206][207] determination, materials property predictions, 208,209 micrograph analysis, 210,211 development of interatomic potentials [212][213][214] and energy functionals 215 to improve materials simulations, and on-the-fly data analysis of high-throughput experiments. 216 …”

Finding New Perovskite Halides via Machine Learning

“…A steady increase in computational power, accompanied by developments in quantum theory and algorithmic breakthroughs that allow for efficient yet accurate quantum mechanical computations, opens the door to computing properties of a wide range of materials that once seemed prohibitively expensive. As a result, high-throughput explorations of the vast chemical space are increasingly being pursued and have significantly aided our intuition and knowledge-base of material properties (Ceder et al, 2011;Jain et al, 2011;Yu and Zunger, 2012;Curtarolo et al, 2013;Pilania et al, 2013Pilania et al, , 2016Sharma et al, 2014;Balachandran et al, 2016;Kim et al, 2016;Mannodi-Kanakkithodi et al, 2016). Massive open source databases of materials properties (including electronic structure, thermodynamic, and structural properties) are now available on the web (Curtarolo et al, 2012;Computational Materials Repository, 2015;Materials Project -A Materials Genome Approach, 2015).…”

Stratified construction of neural network based interatomic models for multicomponent materials

“…[22][23][24][25][26][27][28][29][30][31][32][33]. Artificial neural networks (NNs) have proven to be effective machine learning tools for dealing with multidimensional classification, control, and interpolation problems in various fields.…”