wayground logo

Free Printable Worksheets

Font size

S
M
L
XL
Worksheets

AI Arena - Round1

Total questions: 15

Worksheet time: 8mins

Name
Class
Date
1.

Aanya is working on a project where she needs to analyze a large dataset with many features. To simplify her analysis and improve the performance of her machine learning model, she decides to use a technique for dimensionality reduction. Which of the following algorithms should she choose?

a)

Random Forest

b)

PCA

c)

Gradient Boosting

d)

KNN

2.

In supervised learning, the algorithm learns from:

a)

Unlabeled Data

b)

Labeled Data

c)

Both A & B

d)

None

3.

The vanishing gradient problem is most common in:

a)

CNNs

b)

RNNs

c)

SVMs

d)

Decision Trees

4.

True or False: NLP models convert text into numerical features using embeddings.

a)

True

b)

False

5.

What does 'epoch' mean in training a neural network?

a)

One complete pass through the entire dataset

b)

One gradient update

c)

A random batch

d)

None

6.

In NLP, 'stemming' and 'lemmatization' are used to:

a)

Detect entities

b)

Reduce words to their root form

c)

Find sentiment

d)

Tokenize

7.

In a machine learning project, Arjun is training a neural network to classify images. He notices that the model performs well on the training data but poorly on the validation data. To improve the model's performance, he decides to implement dropout layers. Dropout layers are mainly used to:

a)

Reduce bias

b)

Prevent overfitting

c)

Increase training speed

d)

Normalize data

8.

True or False: The softmax function outputs values between -1 and 1.

a)

True

b)

False

9.

The loss function used in binary classification is usually:

a)

Mean Squared Error

b)

Cross Entropy

c)

Hinge Loss

d)

Log Loss

10.

In NLP, which model architecture is based on self-attention?

a)

LSTM

b)

Transformer

c)

CNN

d)

RNN

11.

True or False: Overfitting means the model performs well on training but poorly on unseen data.

a)

True

b)

False

12.

What is 'tokenization' in NLP?

a)

Breaking text into words or subwords

b)

Converting numbers to text

c)

Removing stopwords

d)

Translating text

13.

Gradient Descent helps to:

a)

Minimize loss

b)

Maximize accuracy

c)

Find gradients

d)

Split dataset

14.

True or False: CNNs are mainly used for sequence prediction tasks.

a)

True

b)

False

15.

Shaan is working on a machine learning project and needs to choose an activation function for his neural network. He is considering various options but wants to identify which of the following is NOT an activation function.

a)

ReLU

b)

Sigmoid

c)

Softmax

d)

Gradient