Question Bank of Applied Machine Learning-2024-25
Unit-1 Introduction to Machine Learning:
Q.1 Explain the difference between human learning and machine learning, including types of
machine learning and their applications.
Q.2 Briefly explain the concept of feature engineering in machine learning.
Q.3 Identify and describe at least three common applications of machine learning in real-world
scenarios, along with the tools and technologies typically used to implement them.
Q.4 How does the choice of a machine learning algorithm impact the performance of a model, and
what factors should be considered when selecting an algorithm for a specific application?
Q.5 What is the difference between supervised learning and unsupervised learning?
Unit-2 Overview of Probability:
Q.1 Discuss the concepts of probability and its significance in machine learning.
Q.2 What is the Central Limit Theorem, and why is it significant in the context of machine learning
and statistical analysis?
Q.3 What is the role of hypothesis testing in machine learning, and how does it impact model
selection and evaluation?
Q.4 Describe the purpose of cross-validation in model evaluation.
Q.5 Explain the concept of a random variable in probability.
Unit-3 Bayesian Concept Learning:
Q.1 Explain Bayes’ theorem and its application in concept learning.
Q.2 What is a Bayesian Belief Network, and how does it facilitate the representation and reasoning
of uncertain knowledge in machine learning?
Q.3 Discuss the importance of Bayesian methods in handling uncertainty in machine learning and
provide an example of a practical application.
Q.4 How do prior and posterior probabilities relate in the context of Bayes' theorem, and why are
they critical for updating beliefs in a Bayesian framework?
Q.5 Below given is a table of Criteria for going to a concert along with
Input value(X) and Weights(W). Make the final decision whether I will
be able to go to the concert or not.
Consider the threshold value as T=1.5
Criteria Input = X Weight = W
Artist is good X1 = 1 W1 = 0.7
Weather is good X2 = 0 W2 = 0.6
Friend will come X3 = 1 W3 = 0.5
Food is served X4 = 0 W4 = 0.3
Beverages are served X5 = 1 W5 = 0.4
Unit-4 Classification and Regression:
Q.1 Compare and contrast supervised and unsupervised learning, providing examples of
classification and regression.
Q.2 Define overfitting in the context of machine learning.
Q.3 What is a confusion matrix, and why is it important in classification tasks?
Q.4 Explain the concept of logistic regression and how it differs from linear regression. In what
scenarios would you choose logistic regression over linear regression?
Q.5 What are the main types of clustering techniques in unsupervised learning, and how do they
differ in terms of their approach and applications?
Unit-5 Neural Networks:
Q.1 What is the purpose of initialization in training a neural network, and why is it important for the
learning process?
Q.2 Describe the architecture and functioning of neural networks, including the concepts of
perceptron learning and backpropagation.
Q.3 What is the difference between Maximum Likelihood Estimation (MLE) and Maximum A
Posteriori (MAP) estimation in the context of parameter estimation for neural networks?
Q.4 What is the function of backpropagation in neural networks?
Q.5 Consider a single perceptron with sign activation function. The
perceptron is represented by weight [0.4 -0.3 0.1] and bias value is 0.
If input vector to the perceptron is X=[0.2 0.6 0.5], then the output of perceptron will be?
Unit-6 Foundations of neural networks and deep learning, Techniques to improve neural
networks:
Q.1 What are the key concepts that form the foundation of neural networks and deep learning?
Q.2 How do regularization techniques help improve neural networks, and what are some common
methods used?
Q.3 What are Recurrent Neural Networks (RNNs), and in what scenarios are they particularly useful
compared to other types of neural networks?
Q.4 What is the role of hyperparameter tuning in machine learning?
Q.5 Explain the architecture and applications of Convolutional Neural Networks (CNNs).
Unit-7 Generative Adversarial Networks
Q.1 What are Generative Adversarial Networks (GANs), and how do they work?
Q.2 What distinguishes model-free from model-based approaches in deep reinforcement learning?
Q.3 What are adversarial attacks, and how do they affect machine learning models?
Q.4 How can GANs be utilized for practical applications like image generation?
Q.5 What techniques can be used to defend against adversarial attacks in deep learning?
