ML Study Hub

Home / Past papers / MLT Quiz 2 / this paper

IIT M DIPLOMA AN4 EXAM QDD4 03 Dec 2023 - 03 Dec 2023

MLT Quiz 214 questions50 marks8 SA · 3 MSQ · 3 MCQDiploma / QDD4
Elapsed 00:00
Comprehension passage - the questions below it refer to this
question figure
Q1SA4 marks
question figure
Q2MSQ4 marksselect all that apply
question figure
  1. The kernel regression is parametric
  2. The kernel regression is non-parametric
  3. The Kernel matrix K is positive semi-definite
  4. In general, Kernel regression is computationally expensive than a simple linearregression
Q3SA4 marks
question figure
Q4MCQ4 marks
question figure
  1. (1.05, 1.20)
  2. (0.85, 0.95)
  3. (1,1)
  4. (3, 3.5)
Q5MSQ4 marksselect all that apply
Which of the following is/are the primary advantages of using L1 regularization
  1. L1 regularization reduces the risk of overfitting.
  2. L1 regularization tends to produce sparse models.
  3. L1 regularization always improves the model’s predictive accuracy on largedatasets.
  4. L1 regularization is primarily used to increase model complexity.
Q6MSQ4 marksselect all that apply
Select all the statements that are true about decision trees and k-Nearest Neighbors (k-NN) in machine learning:
  1. Decision trees are a supervised learning algorithm used for classifications.
  2. The k-NN algorithm is a lazy learner, which means it doesn’t build an explicitmodel during the training phase.
  3. In k-NN, the value of k represents the number of features used forclassification.
  4. k-Nearest Neighbors (k-NN) is a parametric model that requires estimatingprobability distributions.
  5. The depth of the tree is a hyperparameter and is typically chosen using cross-validation.
Q7MCQ4 marks
question figure
  1. w1 → Lasso regression, w2 → Linear regression, w3 → Ridge regression
  2. w1→ Ridge regression, w2 → Lasso regression, w3 → Linear regression
  3. w1 → Linear regression, w2 → Ridge regression, w3 → Lasso regression
  4. w1 → Ridge regression, w2 → Linear regression, w3 → Lasso regression
Q8MCQ4 marks
question figure
  1. β = [0.75, 0.75]
  2. β = [1, 0.5]
  3. β = [0.5, 1]
  4. β = [0.85, 0.12]
  5. None of these
Q9SA4 marks
You are working on a decision tree algorithm to classify whether a bank loan applicant will default on their loan based on several financial factors. The dataset includes Credit Score, Annual Income, Loan Amount, Loan Term as features. The target variable is binary: 1 for ”Default” and 0 for ”No Default.” You have a dataset of 500 loan applicants, and you want to construct a decision tree to predict loan default. To determine the first split (root node), you'll use the information gain as the criterion. Here’s the distribution of loan default in the dataset: Default: 150 applicants No Default: 350 applicants Calculate the Entropy for the initial dataset.
Q10SA4 marks
question figure
Q11SA2 marks
Given a training dataset with 100 data points, how many distances would we have to compute in the process of predicting the label of test-point in the k-NN algorithm with k = 5
Q12SA2 marks
If the proportion of points belonging to class 1 in a node is p, for what value of p is the node’s entropy maximum?
Comprehension passage - the questions below it refer to this
question figure
Q13SA3 marks
question figure
Q14SA3 marks
What will be the predicted label according to the Naive Bayes decision rule?

Answer key

  1. Q1 (SA, 4 marks): 0
  2. Q2 (MSQ, 4 marks): B, C, D
  3. Q3 (SA, 4 marks): 39 to 40
  4. Q4 (MCQ, 4 marks): A
  5. Q5 (MSQ, 4 marks): A, B
  6. Q6 (MSQ, 4 marks): A, B, E
  7. Q7 (MCQ, 4 marks): A
  8. Q8 (MCQ, 4 marks): D
  9. Q9 (SA, 4 marks): 0.83 to 0.92
  10. Q10 (SA, 4 marks): 8191
  11. Q11 (SA, 2 marks): 100
  12. Q12 (SA, 2 marks): 0.5
  13. Q13 (SA, 3 marks): 0
  14. Q14 (SA, 3 marks): 1