Recent advances in machine learning and deep learning-enabled studies on transition metal dichalcogenides

Abstract The machine learning and deep learning (ML/DL) techniques have significantly advanced the understanding and utilization of transition metal dichalcogenides (TMDs) by enabling efficient analysis, prediction, and optimization of their properties. ML/DL methods permit rapid screening, optimization and analysis of two-dimensional (2D) material candidates, potentially accelerating the discovery and development of TMDs with desired electronic, optoelectronic, and energy storage properties. This review provides a comprehensive review of ML/DL methods to enhance 2D materials research via the optimization of synthesis conditions, interpretation of complex data sets, and the use of generative adversarial networks and variational autoencoders for innovative material design and image processing tasks. Furthermore, it highlights the potential of ML/DL techniques in predicting and tailoring the electronic, optical, and mechanical properties of 2D materials to meet specific application requirements.