Figure 1.

Flowchart of active learning. This figure gives an overview of steps involved in the active learning methodology, which are as follows: generate the dataset to train the machine learning model, query the sample space using the relevant acquisition function, select the suitable data point as per the objective function, perform CST Microwave Studio simulation, add the data with its result to the training set, and retrain the machine learning model. The possibility of substitution of CST simulation with a surrogate trained model is demonstrated by the grey shaded box in the left side of the flowchart.