Neural Network and Fuzzy Logic
Neural Network and Fuzzy Logic
(b) What do you mean by a learning rule and what are its types? 8M
(b) If the net input is 0.50, calculate its output when the activation function used is (i) binary
Q:9(a) Explain the procedure for encoding and retrieving associations in BAM . 8M
*** * *
* * *
*** * * *
* * *
*** * *
Pattern(E) Pattern(H)
(1,-1) (-1,1)
(ii) Obtain the response of the net with E and H as input. 10M
Q: 11(a) What do you mean by simple cells and complex cells of a Neocognitron net? 7M
(b) How is electro optical matrix multipliers used to achieve more speed of operation? 8M
Q: 12 (a) What is the function of threshold device used in optical image Recognition system?
8M
(b) How is an ART net designed for both stability and plasticity? 7M
Q: 14 Determine alpha-level sets, strong alpha level sets, height and supports of the
following fuzzy sets with alpha cut= 0.4
Ã0= {(3, .1),(4,.2),(5,.3),(6,.4),(7,.4),(8,.8),(10,1)}
Ã1={(2,.4),(3,.6),(4,.8),(5,1),(6,.8),(7,.6),(8,1)}
Ã2={(2,.4),(4,.8),(5,1),(8,.6)} 15M
Q:15(a) Explain how will you perform union, intersection and compliment operation in fuzzy
(b How are electro optical matrix multipliers used to achieve more speed of operation? 6M
Q.19 (a) Consider the following full CPN using input pair X=(1,1) Y=(0,1). Perform first phase of
training (one step only). Find the activation of the cluster layer units and update the weights using
learning rates of 0.3.
0.2 0.1
X1 Z1 Y1
0.4 0.6
0.3 0.3
0.5 0.7 9M
Y2
X2 Z2
Q. 20(a) Derive the learning rule used in the back propagation net. 7M
(b) Implement a back propagation net for the following information of the given net
2 1 0
0 3 1
Target output is 0.9. Use learning rate as 0.3 and use binary sigmoid activation function. Here W
represents the connection between hidden layer and output layer. W 0 is the bias provided to the output unit
‘y’. V matrix represents connection weights between input and hidden units? 8M