You are on page 1of 2

Question : 2

widrow & hoff learning law is special case of?


a) hebb learning law
b) perceptron learning law
c) delta learning law
d) none of the mentioned
Ans: C widrow & hoff learning law is special case of delta learning law Rule value
Explanation: Output function in this law is assumed to be linear , all other things same.

widrow & hoff learning It comes under the category of Supervised learning, independent of the
activation function of the neuron .it is mainly used to ƒMinimize the squared error between the
desired output value and the neuron active value – Sometimes called LMS (Least Mean Square)
learning rule ƒ

The learning signal r is:

Considered a special case of the delta learning rule

For this case Where Activation function is f(net) = net

After differentiation f ' (net) = 1

Windrof & hoff Learning = ( di- f(net)

Equation is defined as

Delta rule (R) = ( di- f(net ) * f ' (net)

= ( di- f(net )* 1

= Windrof & hoff Learning*1

Delta Rule = Windrof & hoff Learning


Question 5 marks
Q-1) Explain Fuzzy propositions, under the mention connectives,
propositions?

P: Mary efficient, T(P)= 0.8


Q: Ram is efficient, T(Q)= 0.55
connective statements
(a) P: Mary is not efficient
(b) P ^ Q : mary is efficient and so in Ram
(c) P v Q: Either Mary or Ram is efficient
(d) P Q If mary is efficient then so is Ram

Ans :
(i) T(P)= 1- T(P)
= 1-0.8
= 0.2

(ii) T(P^Q) = min (T(P), T (Q))


= min ( 0.8, 0.55)
= 0.55

(iii) T(P v Q = Max (T(P),T(Q)


= max (0.8 , 0.55)
= 0.8
(iv) P Q = max (1- T(P), T (Q))
= max (0.2, 0.55)
= 0.55

You might also like