HOMEWORK  (Machine Learning)

Q-1: briefly explain each of the following concepts:
  1. Learning
  2. Supervised learning
  3. Unsupervised learning
  4. Reinforcement learning
  5. Decision trees
  6. Overfitting and underfitting
  7. Entropy
  8. Information gain
  9. Linear regression
  10. Neural networks

Q-2: Consider the following data set comprised of three binary input attributes (Al, A2, and A3) and one binary output:

dt

Use the decision-tree learning algorithm to learn a decision tree for these data. Show the computations made to determine the attribute to split at each node.


Q-3: Given the following set of training examples: {(-2,-1), (1,1), (3,2)}, find the weights, w0 and w1, for the univariate linear function h(x) = w1*x+w0 that minimizes the squared loss function.

Q-4: Consider the following network of perceptrons which use hard threshold functions for activation. Give the weights for the links and the thresholds for the hidden unit and the output unit such that the network represents the XOR function.

nn_xor

Q-5 (extra credit): The ML repository contains many data sets from various domains. Select one data set from the repository, and apply one of the ML algorithms to it.