Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
4 views

Assignment - Week 2 - Final

Assignment

Uploaded by

tomerfai
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Assignment - Week 2 - Final

Assignment

Uploaded by

tomerfai
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Machine Learning (ML) in Hindi

Week-2 Assignment
Total points: 15
Topics covered:
 Review of Probability Theory
 Linear Regression

Instructions
 Each question is a 1-point multiple choice question (MCQ) with only one correct
answer unless specified otherwise. In case no option matches your (numeric)
answer, select the closest approximation of the same.
 Multiple select questions (MSQ) are 2-point questions with one or more than one
answers correct

1. Which of the following represents a system of linear equations in multiple linear


regression?
a) y = 𝜃0 + 𝜃1 𝑥 + 𝜀
b) y = 𝜃0 + 𝜃1 𝑥1 + 𝜃2 𝑥2 + ⋯ + 𝜃𝑝−1 𝑥𝑝−1 + 𝜀
c) y = 𝜃0 + 𝜀
d) None of the above
Answer: b)

2. In the matrix notation for multiple linear regression, what does the term X represent in
𝐲 = 𝐗𝛉 + 𝛆?
a) The response variable
b) The design matrix
c) The error term
d) The parameter vector
Answer: b) The design matrix

3. What does the term θ0 represent in the linear regression equation y = 𝜃0 + 𝜃1 𝑥 + 𝜀?


a) Weight parameter
b) Bias parameter/Intercept
c) Error term
d) Slope of the line
Answer: b) Bias parameter/Intercept

4. According to the Maximum Likelihood Estimation (MLE) approach for solving linear
regression using the statistical approach, what objective function is minimized?
a) min ∑(yi - ŷi)2
b) min ∑|yi - ŷi|
c) min E[e]
d) min ∑ei
Answer: a) min ∑(yi - ŷi)2

5. What is the area under the curve for a probability density function (pdf) of a continuous
random variable?
a) 0
b) 1
c) ∞
d) -1
Answer: b) 1

6. What is the mean of a uniformly distributed random variable defined over the interval [x 1,
x2]?
a) (x1 + x2) / 2
b) (x1 - x2) / 2
c) (x2 - x1) / 2
d) x1 + x2
Answer: a) (x1 + x2) / 2

7. What is the variance of an exponentially distributed random variable with probability


density function given by 𝑓𝑥 (𝑥) = 𝜆𝑒 −𝜆𝑥 𝑓𝑜𝑟 𝑥 ≥ 0 and zero otherwise?
a) 1/λ
b) 1/λ2
c) λ
d) λ2
Answer: b) 1/λ2

8. According to the Central Limit Theorem, what is the limiting distribution of the normalized
sum of a large number of independent random variables?
a) Uniform distribution
b) Exponential distribution
c) Normal distribution
d) Poisson distribution
Answer: c) Normal distribution

9. [MSQ] Which of the following conditions are true for a positive semi-definite covariance
matrix?
a) All the eigenvalues are non-negative
b) It is a symmetric matrix
c) It must be an orthogonal matrix
d) None of the above is correct
Answer: a) All the eigenvalues are non-negative and b) It is a symmetric matrix
10. What is the purpose of linear regression?
a) To classify data points into different categories
b) To learn the relationship between dependent and independent variables
c) To cluster data points into different groups
d) To reduce the dimensionality of the data
Answer: b) To learn the relationship between dependent and independent variables

11. What is the assumption of homoscedasticity in linear regression?


a) The variance of the errors (residuals) is equal for all values of the independent
variables
b) The variance of the errors (residuals) is not required to be equal for all values of the
independent variables
c) All the independent variables are zero mean
d) All the independent variables are normally distributed
Answer: a) The variance of the errors (residuals) is equal for all values of the
independent variables

12. In linear regression, what does the term 𝜀 represent in the equation: y = 𝜃0 + 𝜃1 𝑥 + 𝜀 ?
a) Response variable
b) Independent variable
c) Residual/Error
d) Parameter to be learned
Answer: c) Residual/Error

13. What is the purpose of the test data or hold-out set in linear regression?
a) To train the model
b) To evaluate the performance of the trained model
c) To select the best features
d) To normalize the data
Answer: b) To evaluate the performance of the trained model

14. Which assumption in linear regression states that the independent variables should not
be highly correlated with each other?
a) Normality
b) Independence
c) Homoscedasticity
d) No multi-collinearity
Answer: d) No multi-collinearity

You might also like