Q-1: briefly explain each of the following concepts:
Learning
Supervised learning
Unsupervised learning
Reinforcement learning
Decision trees
Overfitting and underfitting
Entropy
Information gain
Linear regression
Neural networks
Q-2: Consider the following data set comprised of three binary input attributes (Al, A2, and A3) and one binary output:
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.
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.