AL-405 Machine Learning - Jun 2023 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 6 Marks
   aList and Explain perspectives and issues of Machine Learning.
   How to answer

   • Define machine learning and need for perspectives
   • Explain perspectives e.g. statistical, computational, AI/knowledge-based
   • Discuss issues e.g. data, generalization, overfitting, representation, computation
2. 8 Marks
   bDistinguish between supervised learning and Unsupervised learning with example.
   How to answer

   • Define supervised learning with labelled data example
   • Define unsupervised learning with unlabelled data example
   • Compare on basis of data, goal, algorithms
   • Give examples e.g. classification vs clustering
3. 8 Marks
   aDefine Back propagation and write an algorithm for Back Propagation with examples.
   How to answer

   • Define backpropagation as gradient-based weight update
   • Give forward pass and error calculation steps
   • Give backward error propagation and weight update rule
   • Illustrate with small network example
4. 6 Marks
   bWhat is a Perceptron? Explain the working of a perceptron with a neat diagram.
   How to answer

   • Define perceptron as single-layer binary classifier
   • Explain weighted sum, activation, update rule
   • Describe working steps with example
   • Provide neat labelled diagram
5. 8 Marks
   aExplain how Support Vector Machine can be used for classification of linearly separable data.
   How to answer

   • Define SVM and maximum margin hyperplane
   • Explain support vectors and margin maximization
   • Explain optimization for linearly separable case
   • Give classification example with diagram
6. 6 Marks
   bDefine Decision tree. Explain Decision tree algorithm with example.
   How to answer

   • Define decision tree nodes, branches, leaves
   • Explain splitting criteria e.g. entropy, information gain
   • Outline ID3/C4.5 algorithm steps
   • Illustrate with example dataset
7. 10 Marks
   aUse K Means clustering to cluster the following data in to two groups. Assume cluster centroid are $m1 = 2$ and $m2 = 4$. The distance function used is Euclidean distance $\{2, 4, 10, 12, 3, 20, 30, 11, 25\}$.
   How to answer

   • State initial centroids m1=2, m2=4 and Euclidean distance
   • Show assignment iteration for data points
   • Show recomputation of centroids until convergence
   • List final two clusters
8. 4 Marks
   bExplain Expectation Maximization algorithm with example. And also explain why we need it?
   How to answer

   • Define EM for incomplete/latent variable problems
   • Explain E-step expectation and M-step maximization
   • Give example e.g. Gaussian mixture
   • Explain need for missing data and maximum likelihood
9. 6 Marks
   aExplain hypothesis testing with examples.
   How to answer

   • Define null and alternative hypothesis
   • Explain test statistic, p-value, significance level
   • Outline steps of hypothesis testing
   • Give example e.g. mean test
10. 8 Marks
    bExplain resample methods of machine learning.
    How to answer

    • Define resampling and its purpose
    • Explain cross-validation, bootstrap, etc.
    • Compare methods and use cases
    • Give machine learning examples
11. 8 Marks
    aWhat is Linear Regression? Explain in detail with example and list all the assumptions to be met before starting with Linear Regression.
    How to answer

    • Define linear regression model y = mx + c
    • Explain least squares fitting with example
    • List assumptions: linearity, independence, homoscedasticity, normality
    • Discuss interpretation and limitations
12. 6 Marks
    bDifferentiate between regression and classifications.
    How to answer

    • Define regression for continuous output
    • Define classification for discrete labels
    • Compare on output type, algorithms, evaluation
    • Give examples of each
13. 8 Marks
    aWhat is Gaussian Mixture density estimation with example.
    How to answer

    • Define mixture of Gaussians for density estimation
    • Explain parameters weights, means, covariances
    • Explain fitting via EM with example
    • Give application e.g. clustering
14. 6 Marks
    bExplain different algorithms for dimension reduction with example.
    How to answer

    • Define dimensionality reduction need
    • Explain PCA, LDA, t-SNE etc.
    • Compare algorithms and steps
    • Give example applications
15. 14 Marks
    write short notes on the following:

    1. a)Cross validation
    2. b)Training and validation
    3. c)Cluster
    4. d)Factors.
    How to answer

    • Define cross-validation and k-fold procedure
    • Explain training vs validation split purpose
    • Define cluster in unsupervised learning
    • Explain factors in ML context