Fig. 1 | Scientific Reports

Fig. 1

From: Performance of uncertainty-based active learning for efficient approximation of black-box functions in materials science

Fig. 1

Flow of uncertainty-based active learning. First, a Gaussian process regression (GPR) model is trained by the initial datapoints. Next, the datapoint with the highest uncertainty calculated by GPR is selected, and the property is measured experimentally. By iterating the training of the GPR model, selection of the most uncertain point, and experiments, an efficient approximation of BBFs can be achieved.

Back to article page