Q5. Boolean Perceptron

#1

Consider two perceptrons one denoting boolean AND and one denoting boolean OR. What are two ways to go from an AND perceptron to an OR perceptron?

a) Increase the magnitude of the bias

b) Decrease the value of weights

c) Increase the value of weights

d) Decrease the magnitude of the bias

Decreasing the bias will result in going towards 0 or -ve side of the output whereas on increasing the bias will tend to move the output more towards 0.5 or higher, thus by increasing the weight we will be able to convert AND to OR i.e. cases where one of the weights is TRUE the function should also return TRUE, in case of AND it is less than 0.5 thus by increasing the weight we can turn it into TRUE i.e. option “a)” should be correct not “d)”.

#2

Q8. Dimension Modification 1 We have a training set with 4 data points, as follows : X = (0, 0) => Y = 1 : X = (0, 1) => Y = -1 : X = (1, 0) => Y = -1 : X = (1, 1) => Y = 1. Notice that the data above is not linearly separable, hence the perceptron algorithm will not be able to learn a classifier that gives the correct prediction for all four above data points. Add a 3rd dimension to each of the extra input dimension so that the data becomes linearly separable: a) Third value is equal to first value for each data b) Third value is 1 for one data point, and 0 for other three c) Third value is opposite of second value for each data point d) None of the above I cannot visualize how come just making one point on different plane than the rest of three, there exist a line/plane such that both the classes are separable. Therefore the answer should be None of the above.

#3

Hey @AAYUSHD99, to go from an AND perceptron to an OR perceptron, we need to decrease the bias. Let me attach a figure to explain you :

WhatsApp Image 2021-09-01 at 10.14.09 PM
WhatsApp Image 2021-09-01 at 10.14.09 PM899×1599 106 KB

I guess now it’s self explanatory !

Hope this helps !
Happy Learning :slight_smile:

#4

Sir I think it is W1X1 +W2X2 + bias >= threshold

#5

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.