Midterm 1

Question 1

The model learns the relationship between inputs and outputs by minimizing _____ between predicted and actual values

Answer 1

the difference

Question 2

Supervised learning in regression tasks involves fitting the data to a _____ line using _____ data to predict an output y=h(x) from a given input x.

Answer 2

Straight, labeled

Question 3

The gradient descent algorithm is an optimization method used to minimize a cost function by iteratively updating model parameters in the direction of the _____, which is the _____ of the function

Answer 3

Steepest descent, negative gradient

Question 4

The update step size is controlled by the _____ , and the algorithm continues until convergence, which is typically defined by a sufficiently small change in the cost function

Answer 4

Learning rate

Question 5

(T/F) Logistic regression can only be used for binary classification problems

Answer 5

False

Question 6

(T/F) The output of logistic regression is a probability value between 0 and 1

Answer 6

True

Question 7

(T/F) Logistic regression does not assume any relationship between the input features and the output

Answer 7

False

Question 8

(T/F) Logistic regression uses the sigmoid function to model the probability of a class

Answer 8

True

Question 9

(T/F) SVMs aim to find a decision boundary that maximizes the margin between classes

Answer 9

True

Question 10

(T/F) The SVM cost function can be approximated by piecewise linear functions, though this increases computational complexity

Answer 10

False

Question 11

(T/F) The data points closest to the decision boundary are called support vectors

Answer 11

True

Question 11

(T/F) SVMs achieve better general by maximizing the margin of separation

Answer 11

True

Question 12

(T/F) Overfitting occurs when a model is too complex and captures noise in the training data

Answer 12

True

Question 13

(T/F)A model that overfits will have high training accuracy but poor test accuracy

Answer 13

True

Question 14

(T/F) Overfitting typically happens when the model has too few parameters relative to the training data

Answer 14

False

Question 15

(T/F) Regularization techniques like L1 or L2 can help prevent overfitting by simplifying the model

Answer 15

True

Question 16

(T/F) A model that over fits will perform well on both training and test data

Answer 16

False

Question 17

(T/F) Underfitting typically happens when the model has too many parameters relative to the training data

Answer 17

False

Question 18

(T/F) Underfitting occurs when a model is too simple to capture the underlying patterns in the data

Answer 18

True

Question 19

(T/F) Underfitting occurs when a model learns only the noise in the training data

Answer 19

False

Question 20

(T/F) A model that underfits will have both low training accuracy and low test accuracy

Answer 20

True

Question 21

(T/F) Increasing model complexity can help address underfitting

Answer 21

True

Question 22

(T/F) Regularization techniques add a penalty term to the model cost function to reduce the risk of overfitting

Answer 22

True

Question 23

(T/F) The regularization parameter controls the strength of regularization applied to the model

Answer 23

True

Question 24

(T/F) Small regularization parameters may allow the model to overfit the training data

Answer 24

True

Question 25

(T/F) Too large regularization parameters may result in undercutting by making the model too simple

Answer 25

True

Question 26

(T/F) Regularization can help prevent overfitting even when the training data is small, though more data may be needed for optimal test results

Answer 26

True

Question 27

(T/F) Increasing regularization always improves model performance, regardless of the situation

Answer 27

False

Question 28

(T/F) Removing irrelevant or redundant features can help prevent overfitting and improve model generalization

Answer 28

True

Question 29

(T/F) Adding more training data can reduce overfitting and improve generalization, especially for high-variance models

Answer 29

True

Question 30

(T/F) Adding polynomial features is a good strategy to fix high-variance (overfitting) problems

Answer 30

False

Question 31

(T/F) Reducing the number of features by removing irrelevant ones is a method of reducing the risk of overfitting in a machine learning model

Answer 31

True

Question 32

(T/F) Increasing the regularization parameter(λ) is a method of reducing the risk of overfitting in a machine learning model

Answer 32

True

Question 33

(T/F) Collecting more training data is a method of reducing the risk of overfitting in a machine learning model

Answer 33

True

Question 34

(T/F) Using a more complex model with higher polynomial features is a method of reducing the risk of overfitting in a machine learning model

Answer 34

True

Question 36

Formula for accuracy?

Answer 36

(TP + TN) / (TP + FP + TN + FN)

Question 37

When do we use regression?

Answer 37

When value is continuous, trying to predict value

Question 38

When do we use classification?

Answer 38

When data is labeled, trying to categorize data

Question 39

What is the evaluation metric for regression?

Answer 39

MSE, RMSE, R-Squared

Question 40

What is the evaluation metric for classification?

Answer 40

Accuracy, Precision, Recall, F-1 Score

Question 41

Formula for precision?

Answer 41

TP/(TP + FP)

Question 42

Formula for recall?

Answer 42

TP/(TP + FN)

Question 43

Formula for F1-Score?

Answer 43

2 * [ (Precision * Recall) / (Precision + Recall)