EXTENDED ABSTRACT: The discovery and development of advanced materials is (and has always been) at the heart of technological innovation - driving progress in sectors ranging from energy to aerospace. Traditional materials design approaches, relying heavily on trial-and-error experimentation, are now being disrupted by data-centric methods. Whilst the concept of the materials genome is nearly two decades in the making, it is only recently through machine learning (ML), that we are observing computational design developments - at pace. In this presentation, I will explore the intersection of materials science and artificial intelligence, focusing on how machine learning, integrated with the Materials Genome Engineering (MGE) paradigm, is now reshaping the field. Leveraging ML for materials discovery involves three critical pillars: data curation, algorithmic modeling, and domain knowledge. Recent advances in materials computation and design are discussed, highlighting how data driven approaches may also intersect with theoretical / computational experimentation – allowing datasets for ML exploitation. Some challenges regarding implementing ML in materials science, including data scarcity, interpretability, and the need for standardised data will also be described.
Keywords: big data; materials design; machine learning; artificial intelligence; generative adversarial network