Data Science Supervised Learning Models 5

Question 1

Which activation function is commonly used in hidden layers to avoid vanishing gradients?

Accepted Answer

ReLU

Answer

ReLU keeps positive gradients constant, mitigating the vanishing gradient problem.

Question 2

What does the softmax function produce in a multiclass classifier?

Accepted Answer

A probability distribution over classes

Answer

Softmax normalizes outputs into probabilities that sum to one across all classes.

Question 3

When would you prefer mean absolute error over mean squared error?

Accepted Answer

When you want robustness to outliers

Answer

MAE treats all errors linearly, making it less sensitive to large outliers than MSE.

Question 4

What is the purpose of a confusion matrix?

Accepted Answer

Summarize correct and incorrect predictions by class

Answer

A confusion matrix tabulates true/false positives and negatives for each class.

Question 5

Which scenario most likely indicates overfitting?

Accepted Answer

High training accuracy but low test accuracy

Answer

Overfitting shows strong training performance that fails to generalize to test data.

Data Science Practice Test