wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

QUIZ

Total questions: 45

Worksheet time: 4mins

Name
Class
Date
1.

What is the focus of this Course?

a)

To introduce machine learning concepts

b)

To cover advanced mathematics

c)

To discuss only deep learning

d)

To focus on data visualization

2.

What method is widely employed in machine learning?

a)

Logistic regression

b)

Support vector machines

c)

Decision trees

d)

K-means clustering

3.

What is discussed in relation to deep networks?

a)

Minimization problem of a mathematical function

b)

Data preprocessing techniques

c)

Feature selection methods

d)

Model evaluation metrics

4.

What technique is used to achieve the mathematical goal?

a)

Gradient descent

b)

Random search

c)

Grid search

d)

Bayesian optimization

5.

What is the focus of this week's discussion?

a)

Model training and transfer learning

b)

Data cleaning and preprocessing

c)

Feature engineering techniques

d)

Hyperparameter tuning

6.

What concept is fundamental in natural language processing?

a)

Word embeddings

b)

Image recognition

c)

Time series forecasting

d)

Clustering algorithms

7.

What is the purpose of the Transformer Network?

a)

To be more flexible and robust than RNN

b)

To simplify data processing

c)

To enhance image classification

d)

To reduce computational costs

8.

What is Reinforcement Learning concerned with?

a)

Maximizing rewards in a situation

b)

Minimizing errors in predictions

c)

Optimizing data storage

d)

Improving user experience

9.

What will be discussed after learning the initial steps of Reinforcement Learning?

a)

Q Learning and Deep Q Learning

b)

Supervised learning techniques

c)

Unsupervised learning methods

d)

Data visualization tools

10.

What is the primary goal of machine learning?

a)

To memorize patterns in data

b)

To make predictions or decisions based on data

c)

To create a new programming language

d)

To replace human intelligence

11.

Which of the following is a supervised learning algorithm?

a)

K-Means

b)

Logistic Regression

c)

Apriori

d)

DBSCAN

12.

In logistic regression, the output is:

a)

A continuous value

b)

A discrete class label

c)

A cluster assignment

d)

A probability distribution

13.

What is the role of the activation function in a neural network?

a)

It adds non-linearity to the model

b)

It calculates the loss function

c)

It normalizes the input features

d)

It optimizes the learning rate

14.

Which activation function is commonly used in deep learning

a)

Sigmoid

b)

ReLU

c)

Tanh

d)

Softmax

15.

What does MLP stand for in neural networks

a)

Multi-layered Processor

b)

Multi-Layer Perceptron

c)

Multi-level Processing

d)

Machine Learning Program

16.

The main advantage of deep learning over traditional ML models is

a)

It requires less data

b)

It can automatically extract features

c)

It does not require hyperparameter tuning

d)

It runs faster than all other models

17.

Which of the following is NOT a deep learning framework?

a)

TensorFlow

b)

PyTorch

c)

Scikit-learn

d)

Keras

18.

What is the primary function of backpropagation in neural networks?

a)

To initialize weights

b)

To update weights based on error gradients

c)

To split data into training and testing sets

d)

To prevent overfitting

19.

Which of these is a commonly used optimizer in deep learning?

a)

Gradient Descent

b)

Stochastic Gradient Descent

c)

K-Means

d)

Principal Component Analysis

20.

CNNs are primarily used for

a)

Time-series forecasting

b)

Image processing

c)

Image processing

d)

Text summarization

21.

What is the role of pooling layers in CNNs?

a)

To reduce dimensionality

b)

To detect edges

c)

To normalize data

d)

To improve gradient flow

22.

Which of these pooling methods is commonly used?

a)

Max pooling

b)

Min pooling

c)

Random pooling

d)

Median pooling

23.

A CNN uses filters to

a)

Reduce overfitting

b)

Extract features from images

c)

Increase image resolution

d)

Detect noise

24.

Transfer learning in CNNs refers to

a)

Using a pre-trained model on a new task

b)

Training from scratch

c)

Replacing layers in the model

d)

Fine-tuning hyperparameters

25.

Word embeddings are used to:

a)

Represent words as numerical vectors

b)

Create decision trees

c)

Normalize text data

d)

Translate between languages

26.

Which algorithm is commonly used to generate word embeddings

a)

Naïve Bayes

b)

PCA

c)

Word2Vec

d)

K-Means

27.

What is an RNN primarily used for?

a)

Image classification

b)

Clustering

c)

Sequential data processing

d)

Feature extraction

28.

LSTMs solve which major issue in RNNs?

a)

Slow computation

b)

Vanishing gradients

c)

Lack of parallel processing

d)

Overfitting

29.

Attention mechanisms help in:

a)

Improving learning rate

b)

Assigning importance to input tokens

c)

Reducing overfitting

d)

Removing stopwords

30.

The Transformer model is primarily used in:

a)

Image processing

b)

Text-based tasks

c)

Reinforcement learning

d)

Anomaly detection

31.

What is the key innovation in Transformers?

a)

Convolutional layers

b)

Self-attention mechanism

c)

Principal component analysis

d)

Decision trees

32.

Transformers replace which traditional neural network structure?

a)

CNNs

b)

RNN

c)

GAN

d)

SVMs

33.

The term "sequence-to-sequence" refers to

a)

Converting structured data into images

b)

Mapping an input sequence to an output sequence

c)

Stacking multiple neural networks

d)

Optimizing loss functions

34.

BERT is an example of:

a)

A CNN model

b)

A Transformer-based model

c)

A reinforcement learning algorithm

d)

A clustering method

35.

Reinforcement learning is based on:

a)

Supervised learning

b)

Maximizing cumulative rewards

c)

Feature selection

d)

Naïve Bayes

36.

The key difference between reinforcement learning and supervised learning is:

a)

RL does not require labeled data

b)

RL uses neural networks

c)

RL is slower

d)

RL cannot generalize

37.

Q-learning is an example of:

a)

Supervised learning

b)

Unsupervised learning

c)

Reinforcement learning

d)

Semi-supervised learning

38.

In reinforcement learning, "exploration" refers to

a)

Selecting actions based on past rewards

b)

Trying new actions

c)

Increasing network depth

d)

Fine-tuning hyperparameters

39.

A deep Q-network (DQN) is used for:

a)

Combining deep learning with Q-learning

b)

Image classification

c)

Clustering data

d)
  • Feature extraction

40.

Which of the following statements about bias and variance tradeoff is true?

a)

Increasing bias reduces variance, and vice versa

b)

High bias leads to overfitting

c)

High variance leads to underfitting

41.

In a high-dimensional dataset, which technique is best for avoiding the curse of dimensionality?

a)

Increasing the number of training samples

b)

Reducing the number of features using PCA

c)

Using a more complex model

42.

Which optimization algorithm is most effective when dealing with sparse data and large-scale datasets

a)

Stochastic Gradient Descent (SGD)

b)

Batch Gradient Descent

c)

Newton’s Method

43.

What is the main advantage of using L1 regularization (Lasso) over L2 regularization (Ridge)?

a)

L1 leads to sparse feature selection

b)

L1 is computationally more efficient

c)

L1 reduces overfitting better than L2

d)

L1 does not affect feature weights

44.

The Kernel Trick in SVMs is used to:

a)

Transform non-linearly separable data into a higher-dimensional space

b)

Improve the computational efficiency of linear regression

c)

Reduce model complexity

d)

Minimize bias in the dataset

45.

The No Free Lunch Theorem in ML states that:

a)

Neural networks always outperform traditional ML models

b)

No model is best for all datasets

c)
  • Deep learning works best with small datasets

d)

There exists an optimal universal model