Deep Learning and Machine Learning for Materials Design

The current materials design methodologies depend on AI-driven approaches to accelerate the development of materials. Deep learning (DL) and Machine Learning (ML) play essential roles in advancing materials design by automating analysis tasks and uncovering complex relationships within large datasets. These computational methods offer essential tools for materials innovation. This article investigates the impact of DL and ML on materials design innovation through proposing two taxonomies, i.e., one which focuses on the methods and the other on applications. The method-based taxonomy categorizes the different approaches used in DL and ML for materials design, while the application-based taxonomy presents the real-world scenarios.