You are on page 1of 1

NEURAL NETWORK AND FUZZY LOGIC

HANDS ON TRAINING PROBLEMS

1. Construct an perceptron network for an AND network. Perform 50 epochs of


training.
2. Design and Train a perceptron network for the following problem: Parity:
Consider a 4-input and 1-output problem, where the output should be 'one' if
there are odd number of 1s in the input pattern and 'zero' other-wise.
3. Design and Train a feedforward network for implementing a Full adder.
4. Using the Least Mean square Learning Law design a classifier for the following
Problem:
Class C1 : [-2 2]', [-2 1.5]', [-2 0]', [1 0]' and [3 0]'
Class C2 : [ 1 3]', [3 3]', [1 2]', [3 2]', and [10 0]'
5. Determine the weights of a network with 4 input and 2 output units using
Perceptron Learning Law for the following input-output pairs:
Input: [1101]' [1011]' [0011]' [1111]'
output: [0 0]' [0 1]' [1 0]' [1 1]'
6. Construct a MATLAB Graphical User Interface of back propagation network for
training OR network till the mean square error becomes zero.
7. Write a MATLAB program for developing the following fuzzy membership
functions:
a) Triangular b) Trapezoidal c) Gaussian d) Bell shaped
8. Consider two fuzzy relations of your own and perform max-min composition and
max-product composition
9. Design a Fuzzy logic controller using FIS editor for a Temperature controller.
10. Design a FIS editor using MATLAB GUI toolbox for controlling the water level
and temperature in the boiler using Mamdani model.

You might also like