Human-in-the-loop for Bayesian autonomous materials phase mapping

Autonomous experimentation combines machine learning and laboratory automation to select and perform experiments toward user goals. Accordingly, materials optimization using autonomous experimentation requires fewer experiments and less time than Edisonian studies. Integrating knowledge from theory, simulations, literature, and human intuition into the machine learning model can further increase this advantage. We present a set of methods for integrating human input into an autonomous materials exploration campaign for composition -structure phase mapping. The methods are demonstrated on X-ray diffraction data collected from a thin-film ternary combinatorial library. During the campaign, the user can provide input by indicating potential phase boundaries or phase regions with their uncertainty or indicate regions of interest. The input is then integrated through probabilistic priors, resulting in a probabilistic distribution over potential phase maps given the data, model, and human input. We demonstrate an improvement in phase -mapping performance given appropriate human input.