CS-601 Machine Learning - Jun 2025 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. 8 Marks
   aDefine Machine Learning? Discuss the scope and limitations of Machine Learning.
   How to answer

   • Definition of Machine Learning
   • Scope of Machine Learning with application areas
   • Limitations of Machine Learning
   • Conclusion
2. 6 Marks
   bExplain the role of regression, probability and statistics in Machine Learning.
   How to answer

   • Role of regression in ML
   • Role of probability in ML
   • Role of statistics in ML
   • Summary with examples
3. 7 Marks
   aWhat are Auto Encoders? Explain the types of Auto Encoders?
   How to answer

   • Definition and architecture of Autoencoders
   • Encoder-bottleneck-decoder working
   • Types: vanilla, sparse, denoising, variational, etc.
   • Applications
4. 7 Marks
   bExplain the types of Gradient descent.
   How to answer

   • Gradient descent update rule
   • Types: batch, stochastic, mini-batch
   • Comparison of convergence and computation
   • Advantages and disadvantages
5. 7 Marks
   aExplain the process of implementing CNN in tensor flow.
   How to answer

   • TensorFlow workflow for CNN
   • Data loading, convolution, pooling, dense layers
   • Compilation, training and evaluation steps
   • Example code flow
6. 7 Marks
   bDefine Padding. How does padding works in CNN?
   How to answer

   • Definition of padding
   • Need for padding to preserve spatial size
   • Types: valid vs same, zero padding
   • Effect on convolution output dimensions
7. 6 Marks
   aDiscuss the structure of Long Short Term Memory.
   How to answer

   • LSTM cell structure and gates
   • Forget, input and output gate functions
   • Cell state and hidden state flow
   • Diagram of LSTM unit
8. 8 Marks
   bWhat is Model based and model free learning Q-Learning? Explain.
   How to answer

   • Definition of model-based learning
   • Definition of model-free Q-learning
   • Q-learning update formula
   • Comparison and examples
9. 9 Marks
   aDiscuss the applications of Machine Learning in Speech Processing.
   How to answer

   • ML role in speech recognition and synthesis
   • Feature extraction and acoustic modelling
   • Applications: assistants, transcription, translation
   • Challenges
10. 5 Marks
    bWrite a short note on Tokenization.
    How to answer

    • Definition of tokenization
    • Steps and types: word, subword, sentence
    • Role in NLP pipeline
    • Example
11. 9 Marks
    aExplain the different Data Visualization methods in detail.
    How to answer

    • Need for data visualization
    • Methods: bar, line, scatter, histogram, heatmap, etc.
    • Use cases for each method
    • Tools and examples
12. 5 Marks
    bDiscuss the sigmoid action function in detail.
    How to answer

    • Sigmoid formula and curve
    • Properties: range, derivative
    • Use as activation function
    • Advantages and vanishing gradient limitation
13. 7 Marks
    aExplain the working principle of Markov Decision Process in detail.
    How to answer

    • Definition of Markov Decision Process
    • Components: states, actions, rewards, transitions
    • Markov property and working principle
    • Bellman equations and policy
14. 7 Marks
    bWhat do you mean by dimension reduction? Discuss in detail.
    How to answer

    • Definition of dimensionality reduction
    • Need: curse of dimensionality
    • Techniques: PCA, LDA, t-SNE
    • Applications
15. 4 Marks
    aExplain the following concepts in detail. Natural Language Processing
    How to answer

    • Definition of NLP
    • Components: NLU and NLG
    • Pipeline and applications
    • Challenges
16. 5 Marks
    bExplain the following concepts in detail. Data Preprocessing
    How to answer

    • Definition of data preprocessing
    • Steps: cleaning, integration, transformation, reduction
    • Handling missing values and encoding
    • Importance for ML models
17. 5 Marks
    cExplain the following concepts in detail. Batch normalization
    How to answer

    • Definition of batch normalization
    • Normalization formula per mini-batch
    • Benefits: faster training, regularization
    • Placement in network