Fig. 3 | Scientific Reports

Fig. 3

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

Fig. 3

Scatter plots when predicting \(\:{N}_{\text{v}\text{a}\text{l}}\) data using all the remaining \(\:{N-N}_{\text{v}\text{a}\text{l}}\) data for training (left panels) and the prediction accuracy depending on the iteration steps (right panels) for bandgaps, dielectric constants for electron \(({\varepsilon}_{\text{e}})\) and lattice \(({\varepsilon}_{\text{l}})\) by matminer and orbital field matrix. The ML model is trained by GPR. The 200 independent runs are performed, and the mean and standard deviation are plotted as lines and shaded areas, respectively.

Back to article page