wayground logo

Free Printable Worksheets

Font size

S
M
L
XL
Worksheets

Week4_S2

Total questions: 10

Worksheet time: 5mins

Name
Class
Date
1.

Max pooling with a 2x2 window and stride 2 reduces the output size by:

a)

 Half in each dimension

b)

Quarter in each dimension

c)

No change in size

d)

One-third in each dimension

2.

The output produced by applying a filter to an image is called a (a)   map.

3.

To prevent overfitting, pooling layers reduce the number of (a)   in the network.

4.

What is the main role of pooling layers in a CNN?

a)

Decrease model complexity

b)

Downsample feature maps

c)

Add more parameters to the network

d)

 Improve filter weights directly

5.

Which of the following combinations will most likely lead to overfitting in a CNN?

a)

Small filters + pooling + dropout

b)

Large filters + small dataset + no regularization

c)

Many small filters + large dataset

d)

High learning rate + regularization

6.

The Flatten() layer is used to:

a)

Normalize pixel values

b)

 Reduce the number of features

c)

 Convert feature maps into a vector

d)

Increase batch size

7.

Increasing the number of convolutional filters primarily:

a)

 Reduces model complexity

b)

Replaces pooling layers

c)

Reduces computation time

d)

 Increases model’s feature extraction capacity

8.

Zero-padding is often used to:

a)

Reduce overfitting

b)

 Increase the input image dimension

c)

 Randomly remove neurons

d)

Preserve feature map dimensions

9.

Which activation function is most commonly used in modern CNNs for hidden layers?

a)

Sigmoid

b)

ReLU

c)

Tanh

d)

Softmax

10.

(a)   pooling keeps the maximum value from each region of the input.