Regularization

Question 1

Question 1: What is regularization in machine learning?

Answer 1

A) A technique used to simplify the model to make it easier to interpret.
B) A method used to reduce the generalization error without significantly affecting the training error.
C) A process that increases the number of features in the model to improve accuracy.
D) A technique used to increase the complexity of the model.

Answer: B

Question 2

Question 2: What are the common types of regularization? (Select all that apply)

Answer 2

A) L1 Regularization (Lasso Regression)
B) L2 Regularization (Ridge Regression)
C) Dropout
D) Increasing model training epochs

B) L2 Regularization (Ridge Regr

B) L2 Regularization (Ridge Regr

Answer: A, B & C

Question 3

Question 3: How does L1 regularization (Lasso) differ from L2 regularization (Ridge)?

Answer 3

A) L1 can lead to feature selection by reducing some coefficients to zero.
B) L2 penalizes the square of the coefficients and does not reduce coefficients to zero.
C) L1 increases the model’s ability to fit data perfectly.
D) L2 simplifies the model by completely removing some features.

Answer: A & B

Question 4

Question 4: What is the primary purpose of adding a regularization term to a loss function?

Answer 4

A) To allow the model to perfectly fit the training data.
B) To penalize larger values of the coefficients.
C) To increase the number of parameters in the model.
D) To decrease the training time of the model.

Answer: B

Question 5

Question 5: What impact does regularization have on a model’s training process? (Select all that apply)

Answer 5

A) It can help in preventing the model from overfitting.
B) It may lead to underfitting if the regularization parameter is set too high.
C) It generally increases the complexity of the model to fit better on training data.
D) It helps in reducing the noise within the training data.

Answer: A, B & D

Question 6

Question 6: What are some applications of autoencoders? (Select all that apply)

Answer 6

A) Image coloring
B) Feature extraction and noise reduction
C) Predicting future stock prices
D) Dimensionality reduction

Answer: A, B & D

Question 7

Question 7: Describe the components of an autoencoder. (Select all that apply)

Answer 7

A) Encoder
B) Decoder
C) Pooling layer
D) Code

Answer: A, B, D

Question 8

Question 8: What is the effect of the dropout technique in regularization?

Answer 8

A) It removes certain features entirely from consideration.
B) It randomly ignores specific nodes during training.
C) It increases the number of layers in the neural network.
D) It adjusts the learning rate during training.

Answer:B