MODULE 2 S2.2

Question 1

A constraint where each feature should have as little effect on the outcome as possible (which translates to having a small slope), while still predicting well.

Answer 1

Regularization

Question 2

The ‘slope’ parameter is also called _______ or coefficients.

Answer 2

weight

Question 3

Regression where the coefficients (w) are chosen not only so that they can predict well on the training data, but also to fit an additional constraint.

Answer 3

Ridge Regression

Question 4

T/F Linear Regression is also known as Ordinal Least Squares.

Answer 4

FALSE (Ordinary)

Question 5

It is the sum of the squared differences between the predictions and the true values.

Answer 5

Mean Squared Error

Question 6

It is a linear model for classification problems.

Answer 6

Logistic Regression

Question 7

Logistic Regression happens by fitting a logistic function, also known as the _______________

Answer 7

sigmoid function

Question 8

An alternative to Ridge for regularizing linear regression.

Answer 8

Lasso

Question 9

It is also referred to as w, weights, coefficients.

Answer 9

Slope

Question 10

T/F In Linear Regression, the final output of the model is numeric value (numeric predictions).

Answer 10

TRUE

Question 11

The algorithm used for solving regression problems.

Answer 11

Linear Regression

Question 12

It makes a prediction using a linear function of the input features.

Answer 12

Linear Model

Question 13

T/F In ridge regression, a higher alpha means a more restricted model, so we expect the entries of coef_ to have smaller magnitude for a high value of alpha than for a low value of alpha.

Answer 13

TRUE

Question 14

T/F When comparing training set and test set scores, we find that we predict very accurately on the training set, but the R2 on the test set is much worse. This is a sign of underfitting.

Answer 14

FALSE

Question 15

T/F Ridge regression is a linear regression model that controls complexity to avoid overfitting.

Answer 15

TRUE

Question 16

T/F Dichotomous classes means Yes or No.

Answer 16

TRUE

Question 17

T/F In Ridge regression is α (alpha) is lesser, the penalty becomes larger.

Answer 17

FALSE

Question 18

What regularization is lasso regression?

Answer 18

L1 regularization

Question 19

What does the variable b represent?

Answer 19

Intercept / Offset

Question 20

When comparing training set and test set scores, we find that we predict very accurately on the training set, but the R2 on the test set is much worse. This is a sign of:

Answer 20

Overfitting

Question 21

Linear Regression is also known as _________________

Answer 21

Ordinary Least Squares (OLS)

Question 22

It generates a formula to create a best-fit line to predict unknown values.

Answer 22

Linear Model

Question 23

It means explicitly restricting a model to avoid overfitting.

Answer 23

Regularization

Question 24

It generates a probability between 0 and 1.

Answer 24

Logistic Regression

Question 25

They are called linear because they assume there is a __________ relationship between the __________ variable and each of its ________________.

Answer 25

linear
outcome
predictors

Question 26

T/F The ‘offset’ parameter is also called slope.

Answer 26

FALSE (intercept / b)

Question 27

It is one of the most commonly used alternatives to standard linear regression.

Answer 27

Ridge Regression

Question 28

Types of Linear Models

Answer 28

Linear Regression
Logistic Regression

Question 29

Linear regression finds the parameters ___ and ___ that minimize the ______________ error between predictions and the true regression targets, ___ , on the training set.

Answer 29

w
b
mean squared
y

Question 30

A model that is easy to analyze and understand.

Answer 30

Lasso

Question 31

It is the simplest and most classic linear method for regression.

Answer 31

Linear Regression

Question 32

What regularization is ridge regression?

Answer 32

L2 regularization

Question 33

T/F Ridge is generally preferred over Lasso, but if you want a model that is easy to analyze and understand then use Lasso.

Answer 33

TRUE

Question 34

T/F If α is larger, the penalty becomes larger and w should be smaller.

Answer 34

TRUE