14SCS41
14SCS41
14SCS41
1 a. Define machine learning and discuss various issues in machine learning. (10 Marks)
b. List and explain the different design issues and approaches in machine learning. (10 Marks)
2 a. Illustrate with suitable example the candidate elimination algorithm. (10 Marks)
b. What is decision tree learning? Illustrate ID3 algorithm for simplified version with Boolean
valued functions. (10 Marks)
Important Note : 1. On completing your answers, compulsorily draw diagonal cross lines on the remaining blank pages.
3 a. What are artificial neural networks (ANN). Discuss the different characteristics that are
appropriate for ANN problems. (10 Marks)
b. What is perceptron? Discuss, how a single perceptron can be used to represent the Boolean
functions such as AND, OR (10 Marks)
4 a. Write a note on :
i) Perceptron training rule ii) Gradient descent and delta rule. (10 Marks)
b. Explain stochastic approximation to gradient descent of back propagation algorithm for feed
forward network containing two layers of sigmoideal units. (10 Marks)
5 a. Explain the common operators for Genetic algorithm with an example. (10 Marks)
b. What is Bayesian learning? Discuss the features of Baysian learning methods. (10 Marks)
7 a. What are instance based methods? Explain the key features and disadvantages of these
methods. (10 Marks)
b. Explain the K nearest neighbor algorithm for approximating a discrete valued function
f : Hn V with pseudo code. (10 Marks)
8 a. Compare the purely analytical and purely inductive learning methods. (04 Marks)
b. Write a note on FOIL algorithm. (08 Marks)
c. What is Q learning explain Q learning algorithm assuming deterministic rewards and
actions? (08 Marks)
*****