CS-601 Machine Learning - Jun 2026 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. 7 Marks
   aExplain the scope, limitations and applications of Machine Learning.
   How to answer

   • Define scope of ML: types of learning and tasks
   • Discuss limitations: data, generalization, bias, computation
   • List major applications with examples
   • Conclude with scope vs limitations trade-off
2. 7 Marks
   bDiscuss hypothesis functions, hypothesis testing and data distributions.
   How to answer

   • Define hypothesis function / hypothesis space
   • Explain hypothesis testing steps and errors
   • Describe data distributions and role in learning
   • Relate distributions to generalization
3. 7 Marks
   aDiscuss multilayer neural networks, backpropagation, weight initialization, training and testing.
   How to answer

   • Describe multilayer neural network architecture
   • Explain backpropagation steps and weight update
   • Discuss weight initialization methods
   • Outline training vs testing procedure
4. 7 Marks
   bCompare L1 and L2 regularization. Discuss momentum, hyperparameter tuning and autoencoders.
   How to answer

   • Compare L1 vs L2 penalty, sparsity and formulas
   • Explain momentum in optimization
   • Discuss hyperparameter tuning methods
   • Describe autoencoders structure and use
5. 7 Marks
   aExplain convolution, pooling, dense and loss layers in CNN.
   How to answer

   • Explain convolution layer operation and filters
   • Explain pooling types and purpose
   • Explain dense / fully connected layer
   • Explain loss layers and examples
6. 7 Marks
   bExplain the Inception Network and transfer learning with suitable applications.
   How to answer

   • Describe Inception Network inception modules
   • Define transfer learning concept and steps
   • Give suitable applications for both
   • Highlight benefits and limitations
7. 7 Marks
   aCompare Value Iteration and Policy Iteration with suitable examples.
   How to answer

   • Define Value Iteration update rule
   • Define Policy Iteration: evaluation + improvement
   • Compare convergence, cost with examples
   • Illustrate with grid-world / MDP example
8. 7 Marks
   bExplain Actor-Critic, Q-Learning and SARSA algorithms.
   How to answer

   • Explain Actor-Critic architecture
   • Explain Q-Learning off-policy update
   • Explain SARSA on-policy update
   • Compare exploration and applications
9. 7 Marks
   aExplain data preprocessing, data augmentation, normalization and data visualization in Machine Learning.
   How to answer

   • Define data preprocessing steps
   • Explain augmentation techniques
   • Explain normalization methods
   • Explain visualization role and tools
10. 7 Marks
    bExplain weights, bias, loss function and gradient descent in neural networks.
    How to answer

    • Define weights and bias role
    • Explain loss functions with examples
    • Explain gradient descent steps and variants
    • Relate forward and backward pass
11. 7 Marks
    aDiscuss one-shot learning, dimension reduction and CNN implementation using TensorFlow and Keras.
    How to answer

    • Define one-shot learning and use cases
    • Explain dimension reduction methods
    • Outline CNN implementation in TensorFlow/Keras
    • Mention code flow and layers
12. 7 Marks
    bExplain the Reinforcement Learning framework, Markov Decision Process (MDP) and Bellman equations.
    How to answer

    • Describe RL agent-environment framework
    • Define MDP components: S, A, P, R
    • State Bellman expectation and optimality equations
    • Explain role in policy evaluation
13. 7 Marks
    aDiscuss padding, stride, flattening, subsampling, input channels, and $1\times 1$ convolution.
    How to answer

    • Define padding, stride and effect on size
    • Explain flattening and subsampling
    • Explain input channels concept
    • Explain $1\times 1$ convolution purpose
14. 7 Marks
    bCompare traditional Machine Learning techniques with deep learning for real-world applications.
    How to answer

    • Outline traditional ML features and models
    • Outline deep learning representation learning
    • Compare on data, features, compute, interpretability
    • Give real-world application examples
15. 7 Marks
    aExplain the applications of Machine Learning in computer vision.
    How to answer

    • List computer vision tasks: classification, detection, segmentation
    • Explain ML/DL models used
    • Give application examples
    • Mention challenges
16. 7 Marks
    bExplain the significance of the ImageNet Competition in the development of deep learning.
    How to answer

    • Describe ImageNet Competition / ILSVRC setup
    • Explain breakthrough models: AlexNet etc.
    • Discuss impact on deep learning growth
    • Conclude significance for vision and AI