Machine Learning-guided Design of Emerging 2D-based Materials

A considerable number of novel computational and experimental approaches have emerged in response to the growing interest in two-dimensional (2D) materials. Machine Learning (ML) techniques have become increasingly essential in recent years, with numerous ML-based approaches being employed to investigate diverse classes of materials for specific applications. In this talk, we will highlight two of our recent initiatives to use materials informatics and machine learning in the design of next-generation 2D materials. We developed an ML algorithm to efficiently guide the selection of the lattice parameters for both atomically thin 2D materials and associated heterostructures, with an out-of-sample accuracy as high as 90%. Due to their unique and tunable optoelectronic properties, 2D/organic hybrid materials are intriguing quantum materials, but the development of computational tools and screening of the database of millions of such possible materials remains a challenging task. In the second part of this talk, will present our recent efforts in developing an ML-guided pipeline, materials screening, design, and first-principles calculations of complex quantum materials derived from 2D-based transition metal dichalcogenides and organic molecules via electrochemical intercalation. Using the bootstrap approach, we will show how to efficiently pick interesting materials for future computational and experimental examination based on parameters such as intercalation energy.