Exam MAS-II Practice Question 13

Study guide: Exam MAS-II

A pricing team fits a feed-forward neural network to a quantitative response with p = 6 predictors and a single hidden layer of K = 10 units, each with an activation function g, feeding a linear output layer.

Determine which of the following statements are true.

• I. If g were the identity function, g(z) = z, the fitted network would reduce to a linear model in the six predictors.

• II. The network has 70 weights feeding the hidden layer (including the hidden units' biases) and 11 parameters in the output layer.

• III. The hidden-unit activations are not observed in the data: they are transformations of the predictors whose form is learned during training rather than fixed in advance.

• A) I only

• B) I and II only

• C) I and III only

• D) II and III only

• E) The answer is not given by (A), (B), (C), or (D)

Answer this question and see the worked solution

Concepts tested

• Activation Function
• Hidden Layer
• Neural Network

More questions from this set

• Exam MAS-II Practice Question 12
• Exam MAS-II Practice Question 14