Reducing the Number of Experiments Required for Modelling the Hydrocracking Process with Kriging Through Bayesian Transfer Learning

Abstract The objective is to improve the learning of a regression model hydrocracking process using reduced number observations. When new catalyst used for process, must be fitted. Generating data expensive and therefore it advantageous limit amount generation. Our idea use second set measurements made on an old catalyst. This large enough fit performing models In this work, we knowledge from learn task transfer task. We show that results are greatly improved with Bayesian approach linear kriging model.