DEEP LEARNING WEEK 1

0 .NOTE: For all questions below the output of perceptron is assumed to be 1 and
-1 instead of 1 and 0. wT x < 0 => y = −1
1. Which Boolean function with two inputs x1 and x2 is represented by the following decision
boundary? (Points on boundary or right of the decision boundary to be classified 1)

x2

x1
1 2

(a) AND
(b) OR
(c) XOR
(d) NAND

Answer: (b) OR.

Solution: The line in the diagram is x+y=1 which represents the OR boolean function.
2. Choose the correct input-output pair for the given MP Neuron. (MSQ)
(
1, if x1 + x2 + x3 ≥ 2
y=
0, otherwise

(a) y = 1 for (x1 , x2 , x3 ) = (0, 1, 1)

(b) y = 0 for (x1 , x2 , x3 ) = (0, 0, 1)
(c) y = 1 for (x1 , x2 , x3 ) = (1, 1, 1)
(d) y = 0 for (x1 , x2 , x3 ) = (1, 0, 0)
Answer: (a),(b),(c),(d)
Solution: Putting inputs in the above expression and evaluating gives the answer.
3. Suppose we have a boolean function that takes 4 inputs x1, x2, x3, x4? We have an MP
neuron with parameter θ = 2. For how many inputs will this MP neuron give output y = 1?

a)11
b)21

1
 c)15
d)8

Answer: a)
Solution: Total no of possible boolean inputs is 24 = 16. No of inputs with all 0 is 1 and no
of inputs with three 0 is 4 hence the answer is 16-1-4=11.
4. We are given the following data:

x1 x2 y3
2 4 1
3 -1 -1
5 6 -1
2 0 1
-1 0 1
-2 -2 1

Can you classify every label correctly by training a perceptron algorithm? (assume bias to be
0 while training)

a)Yes
b)No

Answer: a)No
Solution: By plotting x1 and x2 on graph paper we can observe that 1 and −1 can’t be
separated using a line passing through the origin. Hence perceptron will fail to classify all the
points correctly.
5. We are given the following dataset with features as (x1,x2) and y as the label (-1,1). If we
apply the perception algorithm on the following dataset with w initialized as (0,0). What will
be the value of w when the algorithm converges? (Start the algorithm from (2,2))

x1 x2 y
2 2 1
2 -2 1
-2 1 -1

a)(-2,2)
b)(2,1)
c)(2,-1)
d)None of These

Answer: c)(2,-1)

Solution: When we run the perceptron algorithm on our dataset then we observe that all
the points are correctly classified except (-2,1). By doing an update according to the
algorithm w=w-x we get new w =(2,-1). All the points are correctly classified with this w.

2
6. Consider points shown in the picture. The vector w is (-1,0). As per this weight vector, the
Perceptron algorithm will predict which classes for the data points x1 and x2.(MSQ)
y

x1

(2, 1)

w
x

x2 (0, −1) (2, −1)

a)x1=1
b)x2=1
c)x1=-1
d)x2=-1

Answer: b),c)

Solution: The decision boundary is wT x = 0. Hence for w=(-1,0), we get the decision
boundary as y-axis. Anything left to the y-axis (in the direction of w) will have wT x > 0 and
will get labeled 1. A similar argument holds for the left side
7. Given an MP neuron with the inputs as x1,x2,x3,x4,x5 and threshold θ = 3 where x5 is
inhibitory input. For input (1,1,1,0,1) what will be the value of y?

a)y = 0
b)y = 1 since θ ≥ 3
c)y = 1
d)Insufficient information

Answer: a)
Solution: x5 is inhibitory hence, the output is 0 regardless of other inputs.

8. An MP neuron takes two inputs x1 and x2. Its threshold is θ = 0. Select all the boolean
functions this MP neuron may represent. (MSQ)

a)AND
b)NOT
c)OR
d)NOR

3
 Answer: d)
Solution: Only functions with threshold θ = 0 are NOT and NOR. NOT takes only one
input while NOR takes two
 
 with weight vector w = 2 −3 1 and bias b = −2
9. What is the output of a perceptron
when the input is x = 1 0 −1 ?
(a) 0
(b) 1
(c) -1
(d) 2
Answer: (c) -1
Solution: The weighted sum of the input vector and weight vector plus the bias is:

w · x + b = 2(1) + (−3)(0) + 1(−1) + (−2) ≤ 0

Since the output of a perceptron is based on whether the weighted sum is greater than or less
than a threshold value, which is 0, the output of this perceptron is -1.
10. What is the ”winter of AI” referring to in the history of artificial intelligence?

a) The period during winter when AI technologies are least effective due to cold
temperatures.
b) A phase marked by decreased funding and interest in AI research.
c) The season when AI algorithms perform at their peak efficiency.
d) A period characterized by rapid advancements and breakthroughs in AI technologies.

Answer:b)

Solution: Refer to lec 1.2