Exam MAS-II Practice Question 12
Study guide: Exam MAS-II
A neural network for a quantitative response has two inputs, one hidden layer of three units with the ReLU activation g(z) = \max(0, z), and a linear output layer:
f(X) = \beta_0 + \sum_{k=1}^{3}\beta_k\, g\!\left(w_{k0} + w_{k1}X_1 + w_{k2}X_2\right)The fitted weights are:
Unit k |
w_{k0} |
w_{k1} |
w_{k2} |
\beta_k |
|---|---|---|---|---|
| 1 | 0.5 | 1.0 | 0.5 | 0.8 |
| 2 | -1.0 | 0.25 | 1.5 | 0.5 |
| 3 | 0.2 | -0.4 | -1.0 | -1.5 |
and \beta_0 = 0.6.
Calculate the network's prediction for X_1 = 2, X_2 = -1.
- A) 0.47
- B) 0.60
- C) 1.00
- D) 1.50
- E) 1.60
Answer this question and see the worked solution