Fig. 5
From: Machine learning based multi-parameter droplet optimisation model study
Boxplots of the 10 sets of minimum objective function values corresponding to the predicted parameter values generated by the CIJ device using the GP-BO and BO in Loop optimisation methods, respectively. (a) Boxplots of predicted parameter outputs optimised using the GP-BO method for the inkjet device with two objective functions \(\:l\left(x\right)\) and \(\:{l}^{{\prime\:}}\left(x\right)\), and for the microfluidic device with \(\:{l}^{{\prime\:}}\left(x\right)\); (b) Boxplots of predicted parameter outputs optimised using the BO in Loop method for the inkjet device with two objective functions \(\:l\left(x\right)\) and \(\:{l}^{{\prime\:}}\left(x\right)\), and for the microfluidic device with \(\:{l}^{{\prime\:}}\left(x\right)\).
