PART-A
Unit-1
1. Define Concept learning
2. Interpret checkers game as well posed learning problem
3. State the version space representation
4. Write about sampling theory
5. Define hypothesis space search in decision tree learning
6. List out the basic steps for designing a learning system?
7. State decision tree learning algorithm.
8. What is the representational power of perceptrons?
Unit-2
1. Why multilayer networks preferred the certain problems
2. What are some remarks on the back propagation algorithm?
3. Discuss any four applications of ANN?
4. Derive the Gradient Descent Rule.
5. What are Artificial Neural Networks?
6. How to compute expected value and variance of a random variable?
Unit-3
1. State Bayses theorem
2. List out the applications of Reinforcement Learning
3. What is Regularization?
4. Define locally weighted regression.
5. Under what conditions is successful learning possible?
Unit-4
1. What is Genetic Programming?
2. What are Convolution Networks?
3. What is Reinforcement Learning?
4. list out the types of Instances based Learning methods.
5. list out the Models of evaluation and learning?
6. What is PAC Learnability?
7. Define Regularization?
8. List out the applications of Reinforcement Learning?
9. What factors contribute to the popularity of genetic algorithm?
Unit-5
1. Define Genetic Operators.
2. What is Crossover operator?
3. What is Deep Feedforward Network?
4. Difference between analytical and inductive learning methods?
5. What are the limitations of explanation-based learning?
PART-B
Unit-1
1.a) Apply the find –S algorithm to a concept learning task?
b) Explain the perspective and issues in machine learning
2. a) What are the importance of estimating hypothesis accuracy?
b) Discuss version spaces and the candidate elimination algorithm with example (or) Prove Є- exhausting
the version space theorem.
Unit-2
1. a) An illustrative example face recognition in neural networks?
b) Demonstrate general to specific learning with example
2. a) Define perceptron? Explain the representation of the perceptron and its training rule
b) Discuss how a multi-layer network learns using a gradient descent algorithm
3. a) Explain the methods for comparing the accuracy of two hypotheses
b) With suitable example discuss a radial basis function network.
Unit-3
1. a) Explain the concept of Bayes theorem with an example (or) Explain the features of bayesian learning
methods.
b) Discuss derivation of back propagation rule (or) Explain about Back Propagation algorithm.
(or)
Write the steps of ID3Algorithm? And capabilities, limitations of ID3? (or) Contrast the hypothesis space
search in ID3 and candidate elimination algorithm.
2. a) Illustrate the impact of over fitting in a typical application of decision tree learning
b) Discuss Maximum Likelihood and Least Square Error Hypothesis. (or) Discuss the relationship between
the maximum likelihood hypothesis and the least squared error hypothesis.
3. a) Explain about sample complexity for Finite and Infinite Hypothesis Spaces
b) What is Mistake Bound Model of Learning? Explain about it.
Unit-4
1. a) Discuss about genetic Algorithms (or) Describe the representation of hypotheses and genetic
algorithms used in this.
b) Explain about Reinforcement Learning and Q-Learning.
2. a) Differentiate between sequential covering algorithm and FOIL Algorithm (or) Explain about FOIL
algorithm.
b) How rules are post pruned? Explain with an example.
Unit-5
1.a) Explain about learning with perfect domain theories: PROLOG-EBG
b) Explain about Combining Inductive and Analytical Learning (or) ) Distinguish between inductive bias and
estimation bias.
2. a) Explain about the various types of Instances based Learning methods
b) Discuss about using prior knowledge to alter the search objective, and augment search operators.
