Jerng-Sik Song, Chin-An Chang
Journal of Vacuum Science and Technology A: Vacuum, Surfaces and Films
Material discovery has been vastly accelerated due to the data revolution and the application of Optimisation and AI. However, there has not yet exist a method that combines optimisation with sequential decision making to utilise the design of material according to a desired property. In this paper, we propose a multi leader – one follower architecture for material discovery to combine human feedback, surrogate models and reinforcement learning.