Scribd Logo
Upload

Welcome to Scribd!

  • 905 views

    Assignment 8

    Uploaded by

    Vijay
    This document contains the questions and solutions for Assignment 8 in an introduction to machine learning course. There are 7 multiple choice questions covering topics like Bayesian networks, naive Bayes classification, and gradient boosting vs AdaBoost. The questions test concepts around conditional probability calculations, feature independence assumptions, and techniques for reducing overfitting in random forests.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Assignment 8

    Uploaded by

    Vijay
    905 views4 pages
    This document contains the questions and solutions for Assignment 8 in an introduction to machine learning course. There are 7 multiple choice questions covering topics like Bayesian networks, naive Bayes classification, and gradient boosting vs AdaBoost. The questions test concepts around conditional probability calculations, feature independence assumptions, and techniques for reducing overfitting in random forests.

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains the questions and solutions for Assignment 8 in an introduction to machine learning course. There are 7 multiple choice questions covering topics like Bayesian networks, naive Bayes classification, and gradient boosting vs AdaBoost. The questions test concepts around conditional probability calculations, feature independence assumptions, and techniques for reducing overfitting in random forests.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    905 views4 pages

    Assignment 8

    Uploaded by

    Vijay
    This document contains the questions and solutions for Assignment 8 in an introduction to machine learning course. There are 7 multiple choice questions covering topics like Bayesian networks, naive Bayes classification, and gradient boosting vs AdaBoost. The questions test concepts around conditional probability calculations, feature independence assumptions, and techniques for reducing overfitting in random forests.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 4

    Assignment 8

    Introduction to Machine Learning


    Prof. B. Ravindran
    1. (2 marks) The figure below shows a Bayesian Network with 9 variables, all of which are binary.

    Which of the following is/are always true for the above Bayesian Network?
    (a) P (A, B|G) = P (A|G)P (B|G)
    (b) P (A, I) = P (A)P (I)
    (c) P (B, H|E, G) = P (B|E, G)P (H|E, G)
    (d) P (C|B, F ) = P (C|F )
    Sol.(b)
    Refer to the lecture
    2. (2 marks) Consider the following data for 20 budget phones, 30 mid-range phones, and 20
    high-end phones:

    Type Single SIM 5G Comaptability NFC Total


    Budget 15 5 0 20
    Mid-Range 20 20 15 30
    High End 15 15 15 20

    1
    Consider a phone with 2 SIM card slots and NFC but no 5G compatibility. Calculate the
    probabilities of this phone being a budget phone, a mid-range phone, and a high-end phone
    using the Naive Bayes method. The correct ordering of the phone type from the highest to
    the lowest probability is?

    (a) Budget, Mid-Range, High End


    (b) Budget, High End, Mid-Range
    (c) Mid-Range, High End, Budget
    (d) High End, Mid-Range, Budget
    Sol. (c)
    P (Class |x1 , x2 , x3 ) ∼ P (Class) ∗ P (x1 | Class ) ∗ P (x2 | Class) ∗ P ( x3 | Class)
    P( Budget | !SSIM, !5G, NFC) ∼ 20/70 ∗ 5/20 ∗ 15/20 ∗ 0/20 = 0
    P( Mid-range | !SSIM, !5G, NFC) ∼ 30/70 ∗ 10/30 ∗ 10/30 ∗ 15/30 = 0.0238
    P( High-end | !SSIM, !5G, NFC) ∼ 20/70 ∗ 5/20 ∗ 5/20 ∗ 15/20 = 0.0134
    3. (2 marks) Consider the following dataset where outlook, temperature, humidity, and wind are
    independent features, and play is the dependent feature.

    Find the probability that the student will not play given that x = (Outlook=sunny, Tempera-
    ture=66, Humidity=90, Windy=True) using the Naive Bayes method. (Assume the continuous
    features are represented as Gaussian distributions).

    (a) 0.0001367
    (b) 0.0000358
    (c) 0.0000236
    (d) 1

    2
    Sol. (a)

    P(x|no) ∗ P(no) = P(sunny|no) ∗ P(Temperature=66/no) ∗ P(Humidity=90|no) ∗P(True|no)


    ∗ P(no)
    P(x|no) ∗ P(no) = 3/5 ∗ 0.0279 ∗ 0.0381 ∗ 3/5 ∗ 5/14 P(no|x) = 0.0001367
    4. Which among Gradient Boosting and AdaBoost is less susceptible to outliers considering their
    respective loss functions?
    (a) AdaBoost
    (b) Gradient Boost
    (c) On average, both are equally susceptible.
    Sol. (b)
    Gradient Boosting (discussed in the lecture) uses a least squares loss function, while AdaBoost
    uses an exponential loss function. AdaBoost penalizes outliers to an exponential amount,
    whereas Gradient Boost penalizes them to a lesser extent and, thus, cares less about them.
    5. How do you prevent overfitting in random forest models?
    (a) Increasing Tree Depth.
    (b) Increasing the number of variables sampled at each split.
    (c) Increasing the number of trees.
    (d) All of the above.
    Sol. (c)
    Refer to the lecture.
    6. A dataset with two classes is plotted below.

    Does the data satisfy the Naive Bayes assumption?

    3
    (a) Yes
    (b) No
    (c) The given data is insufficient
    (d) None of these
    Sol. (a)
    From the plot, we can infer that the features of the data are independent given the class.
    7. Ensembling in random forest classifier helps in achieving:
    (a) reduction of bias error
    (b) reduction of variance error
    (c) reduction of data dimension
    (d) none of the above
    Sol. (b)
    Refer to the lecture

    You might also like