You are on page 1of 2

Assignment # 4

SAA-20EC2209A
CO4
Q.1) Derive the Perceptron learning weight update.
(Hint: Use PPT- Pattern mapping: Multi layered feedforward Neural Networks, slide number 4 to
10)

Q.2) List and explain the different learning rules used in ANN. (PPT-ANN, Slide
numbers 18-24)
Q.3) Let X be the price of the product and Y be the sales. Assume 10 samples
and design ANN for predicting the price for data in below table. Let initial
weights wi = {w1, …w10}m and bias=1
(Product Price) X Y Sales (RS) Wi
89 1201 0.12
75 550 0.25
55 274 0.55
20 1032 0.66
31 678 0.41
101 901 0.11
154 200 0.87
210 150 0.89
110 801 0.99
15 ? 0.74
(https://www.renom.jp/notebooks/tutorial/beginners_guide/feedforward_exa
mple_1/notebook.html)
Q.4) Implement 2 Layered ANN with Adaline activation at output. Explain
suitable each step with weight update equation.
Q.5) List and explain different firing rate and thresholding used in ANN.
Q.6) The graph below represents a regression line predicting Y from X. The
values on the graph show the residuals for each prediction's value. Use this
information to compute the Sum of Squared Error.

You might also like