1. You are given a dataset with predictor variables X and labels Y. The dataset has the following scatter plot: Y iii) 55 = 50 45 40 35 30 25 Bo + Bx + x 0 0.5 1 1.5 X 2 k 2.5 3 Consider the following three machine learning models, where x is the input variable, is the predicted label value, and Bo, B, B,... are the trainable parameters of the model. (here x is just a number, not a vector) i) = Bo + x ii) Bo + Bx + 3x + 3x + B4 log(x + B5) + Be + 37 tan(x + 38) = 3.5 4 Suppose you train each of these three models using the above dataset. a) Which of these three models will have the lowest training error? Please explain. b) Which model is most likely to have the lowest test error? Please explain. c) Which of the three models is the best model to use for this dataset? Why?