Accelerating materials modeling and discovery with artificial intelligence and machine learning

Materials form the backbone of every society. This is evidenced from the naming of human civilizations as Stone Age, Bronze Age, Iron Age, to name a few. Traditional materials discovery relies on trial and error approaches thereby leading to a design to deploy period of 20- 30 years. To address this challenge, in this talk, we will discuss the application of artificial intelligence (AI) and machine learning (ML) in accelerating materials modeling and discovery. Specifically, three aspects where AI and ML can be used include: (i) data-driven models for materials property predictions, (ii) natural language processing (NLP) for extracting information from the literature, (iii) physics-informed machine learning for materials modeling. To demonstrate these aspects, three problems will be discussed. First focuses on developing interpretable ML models for predicting 25 properties of glasses made of a few among 84 elements of the periodic table. This work covers nearly the entire periodic table for glass forming elements. Second focuses on extracting information on glasses and other materials from literature to answer specific queries. We will also discuss on MatSciBERT, the first materials-aware language model. Third, we will discuss on how to accelerate simulations using physics-informed ML (PIML). Here, we will discuss how interaction laws in nature can be discovered directly from the trajectory of physical systems using PIML. Altogether, the talk will cover various aspects of AI and ML that has been used to accelerate materials discovery. Finally, a brief outlook on the future prospects will be discussed.