Machine Learning in Physics, Chemistry, and Materials Science: Materials Discovery and Design

Nowadays, we are witnessing a tremendous increase in data generation enabled by advances in experimental techniques and theoretical and computational developments. This availability of data, associated with new tools and technologies capable of storing and processing that data, culminated in the so-called data science. One of the most prominent areas (machine learning), which aims to identify correlations and patterns in the data sets. These algorithms have been used for decades in different areas. Only recently the community introduced its application for materials, due to the creation, standardization, and consolidation of consistent databases. The use of these methodologies allows to extract knowledge and insights from the huge amount of raw data and information now available. The area presents several opportunities for solving challenges in physics, chemistry, and materials science. Specifically, machine learning methods are a powerful tool for discovering and designing new materials with desired and optimized properties and functionalities. In this article, we present the context of the emergence of machine learning, its foundations, and applications for the discovery and design of materials.

Keywords
Artificial intelligence; data-driven science; tool-driven science; materials informatics