You are on page 1of 29

All Lectures Quiz’s With Answer

AI with Irfan Malik


Quiz No: 03
1. What is the output of the following python code? a = "Hello, World!"
 print(a.split(","))
 Hello, World!
 ['Hello', ' World!'] (True)
 Hello World!
 {'Hello': 'World!'}

2. What is the output of the following python code?


a=9
if a==10:
print("a")
else:
print("b")
 9
 10
 a
 b (True)

3. What is the main difference between a list and a tuple?


 A list is ordered, while a tuple is unordered.
 A list can be modified, while a tuple cannot be modified once created.
 A list can store multiple data types, while a tuple can only store one data type.
 There is no difference between a list and a tuple.

4. What symbol is used to separate a dictionary's key-value pairs?


 ,
 : (True)
 =
 ;

5. Which of the following options shows the correct syntax to create an empty list in
Python?
 empty_list = {}
 empty_list = [] (True)
 empty_list = ()
 empty_list = ""

Quiz no: 04

1. How are key-value pairs represented in Python dictionaries?


 (key, value)
 [key: value]
 {key, value}
 {key: value} (True)

2. Which operator is used for exponentiation in Python?


 +
 *
 ** (True)
 /

3. Which of the following data structures in Python stores unique elements only?
 Lists
 Tuples
 Sets (True)
 Array

4. What is the purpose of using loops in programming?


 To repeat a set of instructions a specific number of times. (True)
 To execute a set of instructions only if a certain condition is met.
 To store a collection of elements in memory.
 To perform mathematical calculations.

5. What is the primary purpose of an if statement in programming?


 To iterate over a collection of elements.
 To evaluate a condition and execute a block of code based on its result. (True)
 To perform arithmetic operations.
 To define a function or method
Quiz no: 05
1. What is a parameter in a function?
 A variable used to store the function's result
 A value passed to the function when it is called (True)
 A reserved word in Python
 A variable passed to the function when it is defined

2. Which loop is better suited when the number of iterations is known in advance?
 for loop (True)
 while loop
 do-while loop
 They are equally suitable

3. How do you define a function in Python?


 define function_name():
 def function_name(): (True)
 function function_name():
 create function_name():

4. What is the purpose of a function in Python?


 To declare variables
 To perform a specific task (True)
 To create loops
 To import modules

5. What is the purpose of the "return" statement in a function?


 To call the function
 To print a value
 To specify the function's name
 To return a value from the function (True)
Quiz No: 06

1. What is the purpose of an API key in API authentication?


 It encrypts the data sent over the network.
 It authorizes access to the API resources. (True)
 It provides a unique identifier for API requests.
 It compresses the API response data.

2. What is the purpose of the range() function in a for loop?


 It controls the number of iterations.
 It generates a sequence of numbers. (True)
 It checks for a condition to continue the loop.
 It initializes the loop counter variable.

3. Which loop in Python is used for iterating over a sequence, such as a list or
string?
 for loop (True)
 while loop
 repeat loop
 iterate loop

4. Which Python library can be used to interact with OpenAI API for sentiment
analysis?
 openai-python (True)
 openai-client
 openai-python-sdk
 openai

5. What is the purpose of Openai API's sentiment analysis feature?


 To convert text into speech.
 To analyze the sentiment of a given text. (True)
 To translate text into different languages.
 To generate code snippets based on user input.
Quiz No: 07
1. Which type of Machine Learning is used when the training data has input
features and corresponding target labels?
 Supervised Learning (True)
 Unsupervised Learning
 Reinforcement Learning
 Semi-Supervised Learning

2. What is the primary goal of Supervised Learning?


 To maximize reward by interacting with an environment
 To discover hidden patterns or structures in data
 To predict target labels for new input data (True)
 To identify anomalies or outliers in a dataset

3. Which type of data lacks a specific format and does not fit into a traditional
database structure?
 Unstructured data (True)
 Semi-structured data
 Structured data
 Relational data

4. Which type of data may include text documents, images, videos, audio files,
and social media posts?
 Structured data
 Semi-structured data
 Unstructured data (True)
 Hierarchical data

5. What is data preprocessing?


 The process of cleaning and transforming raw data before analysis (True)
 The process of collecting and storing data in a database
 The process of visualizing and predicting data labels
 The process of modeling and predicting future data trends
Quiz no: 08
1. Outliers are data points that:
 Have the highest frequency in the dataset
 Represent missing values in the dataset
 Are used to calculate the standard deviation
 Deviate significantly from other data points in the dataset (True)
 Are excluded from the analysis
 Are always errors in the dataset

2. Which measure of central tendency is not affected by outliers?


 Median (True)
 Mode
 Mean
 Range
 Variance
 Skewness

3. Which of the following is a step in data preprocessing during exploratory data analysis?
 Data cleaning (True)
 Data prediction
 Hypothesis testing
 Model training
 Feature engineering
 Statistical inference

4. What is the measure of central tendency that is affected by outliers?


 Range
 Interquartile Range (IQR)
 Variance
 Standard deviation
 Mean (True)
 Median

5. What does the Interquartile Range (IQR) measure?


 The average of the dataset
 The spread of the dataset
 The range of the middle 50% of the dataset (True)
 The standard deviation of the dataset
 The skewness of the dataset
 The correlation between two variables
Quiz No: 09
1. In box plots, the upper and lower edges of the box represent which values, respectively
 Mean and median
 Median and mode
 Mode and mean
 25th and 75th percentiles (True)
 Maximum and minimum values
 Interquartile range and standard deviation

2. The value that appears most frequently in a dataset is known as:


 Mean
 Median
 Mode (True)
 IQR
 Correlation
 Option 7

3. Which measure of central tendency is the middle value in a sorted dataset?


 Mean
 Median (True)
 Mode
 IQR
 Correlation
 Visualization

4. Which of the following correlation coefficients indicates the strongest linear


relationship between two variables?
 -0.2
 0.1
 0.5
 -0.9 (True)
 0.3
 -0.6

5. Which measure of association determines the strength and direction of a linear


relationship between two variables?
 Mean
 Median
 Mode
 IQR
 Visualization
 Correlation (True)
Quiz No: 10
1. Which of the following is an example of categorical data?
 Temperature
 Age
 Gender (True)
 Sales revenue
 Stock prices
 Customer feedback

2. What is the primary purpose of a heatmap in data visualization?


 Displaying hierarchical relationships
 Showing correlations between variables (True)
 Representing geospatial data
 Visualizing time series data
 Comparing proportions
 Analyzing text sentiment

3. What is the five number summary of a data set?


 Mean, median, mode, range, standard deviation
 Minimum, maximum, median, first quartile, third quartile (True)
 Minimum, first quartile, mean, third quartile, maximum
 Median, interquartile range, variance, skewness, kurtosis
 Mode, range, mean absolute deviation, coefficient of variation, z-score
 None of the above

4. What is the value of the median in the following five number summary?
Min = 12, Q1 = 15, Median = , Q3 = 21, Max = 24
 12
 15
 18 (True)
 21
 24
 None of the above

5. What are the two main types of data in statistics?


 a) Qualitative and quantitative
 b) Discrete and continuous
 c) Nominal and ordinal
 d) Categorical and numerical
 e) Primary and secondary
 Both a) and d) (True)
 Both c) and d)
Quiz No:11
1. What is the goal of a machine learning model during the training phase?
 To minimize the training error
 To maximize the testing accuracy
 To generalize well to new, unseen data (True)
 To overfit the training data
 To achieve a perfect match with the training data

2. What does the term "overfitting" mean in machine learning?


 When a model fails to capture the underlying patterns in the data
 When a model performs well on the training data but poorly on new, unseen
data (True)
 When the training and testing sets have significantly different distributions
 When the model has too few parameters to learn from the data None of the
above

3. Which of the following is a common encoding technique used for handling


categorical variables?
 One-Hot Encoding (True)
 Standardization
 Principal Component Analysis (PCA)
 Mean Imputation
 Ordinal Encoding
 None of the above

4. What is classification in machine learning?


 A process of splitting data into training and testing sets
 Converting categorical variables into numerical form
 Predicting categorical labels or classes for new data (True)
 Transforming continuous variables into discrete values
 None of the above

5. Which of the following methods is commonly used for splitting data into
training and testing sets?
 train_test_split() (True)
 fit_transform()
 cross_val_score()
 predict()
 evaluate()
Quiz No: 12
1. In logistic regression, the dependent variable is:
 Continuous
 Categorical (True)
 Both continuous and categorical
 Discrete
 None of the above

2. In a regression model, which variable is the one you are trying to predict?
 Dependent variable (True)
 Independent variable
 Predictor variable
 Control variable
 None of the above

3. Which of the following statements is correct?


 Dependent variables are also known as features.
 Independent variables are the outcome of the study.
 Dependent variables are pre-processed by the researcher.
 Independent variables are also known as response variables.
 None of the above (True)

4. When splitting a dataset into training and testing sets, what is the purpose of the
testing set?
 To evaluate the model's performance on unseen data (True)
 To train the model
 To validate the model's performance on the training data
 To increase the size of the dataset
 None of the above

5. What does the correlation value of 0.075 indicate?


 A weak positive correlation (True)
 A strong positive correlation
 A weak negative correlation
 A strong negative correlation
 None of the above
 The correlation coefficient value is invalid.
Quiz No: 13
1. In regression analysis, which variable is considered the dependent variable?
 The variable that is manipulated in the study
 The variable that is used as a control in the study
 The variable that is the outcome or response being predicted (True)
 The variable that is used to predict the other variables
 The variable that represents the mean of the dataset
 None of the above.

2. Which data structure in Python uses square brackets to enclose its elements?
 Lists (True)
 Tuples
 Dictionaries
 Sets
 Arrays
 None of the above.

3. How is the IQR calculated in a dataset?


 IQR = (Q3 - Q1) / 2
 IQR = Q3 - Q1 (True)
 IQR = (Q3 + Q1) / 2
 IQR = (Q3 + Q2) / 2
 IQR = Q3 - Q2
 None of the above.

4. What is the primary objective of Reinforcement Learning in Machine Learning?


 To classify data into different categories or classes.
 To discover hidden patterns and structures in unlabeled data.
 To make predictions based on past data.
 To optimize an agent's behavior in an environment to achieve a goal. (True)
 To create a model that can work with streaming data.
 None of the above.

5.In the context of NLP, what are stop words?


 Words with a positive sentiment used for sentiment analysis.
 Words that need to be translated when performing machine translation.
 Words that appear frequently in the text and are usually removed during preprocessing.
(True)
 Words that are often used in Named Entity Recognition tasks.
 Words used for creating word embeddings in language models.
 None of the above.
6. What is the Interquartile Range (IQR) used to measure?
 The spread of data around the mean (True)
 The central tendency of the data
 The strength of the relationship between two variables
 The difference between the maximum and minimum values of the data
 The correlation coefficient of the data
 None of the above.

7. What is the key difference between a list and a tuple in Python?


 Lists are mutable, while tuples are immutable. (True)
 Tuples are mutable, while lists are immutable.
 Lists can store only numeric data types, while tuples can store any data type.
 Tuples have a fixed length, while lists can grow or shrink dynamically.
 Lists are ordered, while tuples are unordered.
 None of the above.

8. What are tokens in NLP?


 Special characters used to represent emotions in text.
 Sequential chunks of characters or words used for language processing. (True)
 Words that have a unique representation in a language model.
 Linguistic structures used in syntactic analysis.
 Single characters in a text, including spaces and punctuation marks.
 None of the above.

9. Which step of text preprocessing involves converting words to their base or root form?
 Tokenization
 Lemmatization (True)
 Stopword Removal
 Lowercasing
 Embedding
 None of the above.

10. What does GPT stand for in the context of language models?
 General Processing Toolkit
 Generic Processing Technology
 Generative Pre-trained Transformer (True)
 Global Parsing Technique
 Generalized Parsing Tokenizer
 None of the above.
Quiz No: 14
1. You are a data analyst working for a real estate company. Your task is to analyze the
prices of houses in a particular neighborhood. You have collected the prices of 11
houses in that area. Let's use this data to answer the following questions related to
Interquartile Range (IQR)
Data: [300, 450, 500, 550, 600, 650, 750, 800, 900, 950, 1000]
What is the third quartile (Q3) of the given data set?
 550
 650
 750
 800
 900 (True)
 950

2. You are a scientist conducting a research study to investigate the effects of different
factors on plant growth. You collect data from several experimental setups, each
representing different conditions for plant growth. Use your knowledge of dependent
and independent variables to answer the following question.
In the research study on plant growth, you measure the height of the plants and
record various environmental conditions. What are the dependent and independent
variables in this study?
 Dependent Variable: Plant Height, Independent Variable: Environmental Conditions
(True)
 Dependent Variable: Environmental Conditions, Independent Variable: Plant Height
 Both Plant Height and Environmental Conditions are Dependent Variables
 Both Plant Height and Environmental Conditions are Independent Variables
 Plant Height is the Dependent Variable, but Environmental Conditions are neither
Dependent nor Independent Variables
 Environmental Conditions is the Dependent Variable, but Plant Height is neither
Dependent nor Independent Variables

3. You are a data scientist working on a research study to identify criminals based on their
history. Your primary goal is to predict the criminals and observe their activities to avoid
any criminal incident in the city. What class label would you assign to criminal and non
criminal. Also you have to identify which model you would use for prediction.
 Negative (Criminal), positive (non-criminal) Linear regression(Model)
 Negative (non-criminal), positive (criminal) Linear regression (Model)
 Negative (non-criminal), positive (criminal) Logistic regression (Model) (True)
 Negative (Criminal), positive (non-criminal) Logistic regression(Model)
4. You are a researcher working on creating an advanced AI system for a robotic arm to
perform complex tasks. The arm will improve gradually. Use your knowledge of machine
learning techniques to answer the following question.
 Supervised Learning
 Unsupervised Learning
 Reinforcement Learning (True)
 Deep Learning
 Transfer Learning
 Semi-Supervised Learning

5. Hassan studies maths and physics. Your job is to find the relationship between his math
and physics performance. He scored grades, and you calculated the correlation
between the two subjects, which exhibits a strong positive correlation. What can you
say about his performance in math and physics exams?
 a) When his math scores increase, his physics scores also increase significantly. (True)
 b) When his math scores increase, his physics scores decrease significantly.
 c) his math and physics scores have a weak relationship.
 d) his math and physics scores are unrelated.
 e) both a and f
 f) He is performing equally in both subjects
 g) He is Performing Bad in both subjects
Quiz No: 15
1. An AI agent learns to play a computer game by trial and error, receiving rewards or
penalties based on its actions. Which type of learning is this?
 a) Reinforcement Learning (True)
 b) Supervised Learning
 c) Unsupervised Learning
 d) None of the above
 Both a) and b)

2. You have developed a binary classification model to predict whether a customer will
make a purchase or not. After evaluating the model on a test dataset, you found that it
correctly predicted 87% of the instances. What is this evaluation metric called?
 Interquartile Range (IQR)
 Mean Square Error
 Accuracy (True)
 Feature Scaling
 None of the above

3. You are comparing the performance of two regression models that predict stock prices.
You want to measure the average squared difference between the predicted values and
the actual values to assess their accuracy. Which evaluation metric should you use?
 a) Interquartile Range (IQR)
 b) Mean Square Error (True)
 c) Accuracy
 d) Feature Scaling
 e) None of the above
 f) Root Mean Square Error
 Both b and f

4. You have a large dataset of customer transactions, and you want to identify groups of
similar customers based on their buying behavior without using any labeled data. Which
type of learning is most appropriate for this task?
 a) Reinforcement Learning
 b) Supervised Learning
 c) Unsupervised Learning (True)
 d) None of the above
 Both a) and c)
5. You are studying the survival rates of patients with a particular medical condition based
on various clinical factors. Your goal is to build a model to predict the probability of
survival for future patients. Which type of regression analysis should you use?
 a) Linear Regression
 b) Correlation
 c) Logistic Regression (True)
 d) None of the above
 Both a) and c)

6. You are analyzing the performance of students in a challenging mathematics


competition. You want to determine if there is a significant linear relationship between
the number of hours they spent studying and their final scores. Which statistical
method should you use to quantify the strength of this relationship?
 a) Linear Regression
 b) Correlation (True)
 c) Logistic Regression
 d) None of the above
 Both a) and b)
Quiz No: 16
1. In a confusion matrix, which cell represents instances that are actually positive but are
incorrectly predicted as negative?
 True Positive (TP)
 False Positive (FP)
 True Negative (TN)
 False Negative (FN) (True)
 Positive Error (PE)
 Negative Error (NE)

2. If a model has a high recall and low precision, what can be said about its performance?
 It has high accuracy
 It performs well on negative class predictions
 It performs well on positive class predictions
 It may be suitable for applications where false negatives are costly (True)
 It may be suitable for applications where false positives are costly
 Its performance is balanced

3. In a binary classification problem, if a model predicts 100 instances as positive and 50 of


them are actually positive, while the other 50 are negative, what is the True Positive (TP)
count?
 100
 50 (True)
 75
 25
 150
 0

4. Accuracy is defined as:


 The ratio of true positives to all positive predictions
 The ratio of true positives to all actual positives
 The ratio of true positives to true negatives
 The ratio of true negatives to all negative predictions
 The ratio of correct predictions to total predictions (True)

5. When the threshold for classifying positive instances is increased, how does it typically
affect precision and recall?
 Precision increases, recall increases
 Precision increases, recall decreases (True)
 Precision decreases, recall increases
 Precision decreases, recall decreases
 Precision stays the same, recall increases
 Precision stays the same, recall decreases
6. If a model classifies all instances as negative, what would be the likely impact on its
accuracy, recall, and precision?
 Accuracy increases, recall decreases, precision decreases (True)
 Accuracy decreases, recall increases, precision increases
 Accuracy decreases, recall decreases, precision decreases
 Accuracy increases, recall increases, precision increases
 Accuracy stays the same, recall decreases, precision increases
 Accuracy stays the same, recall increases, precision decrease
Quiz No: 17.1
1. In a binary classification problem, the precision is calculated as:
 (True Positives + True Negatives) / Total Predicted Positives
 True Positives / (True Positives + False Negatives)
 True Positives / (True Positives + False Positives) (True)
 (True Positives + True Negatives) / Total Predicted Negatives
 (True Positives + False Negatives) / Total Actual Positives
 (True Positives + False Positives) / Total Actual Positives

2. Which of the following metrics penalizes false positives more than false negatives?
 Accuracy
 Precision (True)
 RecallF1-score
 True Positive Rate
 False Positive Rate

3. What does the True Negative Rate represent in a confusion matrix?


 Correctly predicted positive cases
 Incorrectly predicted positive cases
 Correctly predicted negative cases (True)
 Incorrectly predicted negative cases
 The total number of positive cases
 The total number of negative cases

4. Which metric can be particularly useful when the class distribution is imbalanced, and
we want to assess the model's ability to identify the minority class
 Accuracy
 Precision
 Recall (True)
 F1-score
 Specificity
 False Positive Rate

5. If a model predicts all instances as positive, what would be the likely impact on
precision and recall?
 Precision increases, Recall increases
 Precision decreases, Recall increases (True)
 Precision increases, Recall decreases
 Precision decreases, Recall decreases
 Precision remains unchanged, Recall increases
 Precision increases, Recall remains unchanged
6. What is the term used to describe the situation when a model predicts an instance as
negative when it's actually positive?
 False Positive
 False Negative (True)
 True Positive
 True Negative
 Type II Error

7. In a medical diagnosis scenario, which metric would be most critical to minimize false
negatives, ensuring that no actual positives are missed?
 Accuracy
 Precision
 Recall (True)
 F1-score
 True Negative Rate
 False Positive Rate
Quiz No: 17.2
1. Alex has trained a model to predict house prices. He wants to measure difference
between the predicted prices and the actual prices. Which evaluation metric should he
use?
 R-squared
 Mean Absolute Error (True)
 Root Mean Squared Error
 Mean Squared Logarithmic Error
 Mean Squared Error
 Huber Loss

2. Sarah is training a neural network, and she notices that the loss function is oscillating
and not converging. Which hyperparameter should she consider adjusting to potentially
address this issue?
 Activation function
 Number of hidden layers
 Learning rate (True)
 Batch size
 Regularization strength
 Dropout rate

3. John is working on a fraud detection model. He wants to assess the overall correctness
of his model's predictions, accounting for both true positives and true negatives. Which
evaluation metric should he use?
 Precision
 Recall
 F1-score
 Accuracy (True)
 ROC-AUC
 Mean Absolute Error

4. Emily is using a confusion matrix to evaluate a classification model. She wants to


measure the proportion of true positive predictions relative to the total number of
positive instances. Which metric does this correspond to?
 Precision (True)
 Recall
 F1-score
 Specificity
 ROC AUC
 False Positive Rate
Quiz No: 18
1. Which of the following is NOT a common value for the learning rate?
 0.01
 0.001
 0.1
 0.0001
 1.5 (True)
 0.00001

2. What is the purpose of adjusting the batch size during training?


 To determine the number of hidden layers in the network
 To control the learning rate of the optimizer
 To change the architecture of the neural network
 To balance between computational efficiency and model convergence (True)
 To change the activation function of the network
 To initialize the loss function

3. What is the purpose of the loss function in neural networks?


 To control the number of layers in the network
 To define the structure of the neural network architecture
 To initialize the weights and biases
 To regulate the learning rate during training
 To measure the difference between predicted and actual values (True)
 To determine the batch size for training

4. What are hyperparameters in the context of neural networks?


 The weights and biases of the neural network
 Parameters that are set before training and control the learning process (True)
 The actual data points used for training
 The predicted values generated by the neural network
 The gradient values computed during backpropagation
 The number of hidden layers in the neural network

5. What is the input size typically used for the MNIST dataset?
 28x28 mm
 32x32 mm
 64x64 mm
 28x28 pixels (True)
 16x16 mm
 20x20 pixels
6. Which of the following hyperparameters determines the step size taken during gradient
descent optimization?
Activation function
 Loss function
 Batch size
 Learning rate (True)
 Number of epochs
 Number of neurons in each layer

7. What is the softmax function used for in neural networks?


 Initializing the weights and biases
 Computing the gradients during backpropagation
 Preventing overfitting in the network
 Selecting the optimal learning rate
 Converting raw scores into values for decision making in multiple classes (True)
 Regularizing the weights of the network
Quiz No: 19
1. Which of the following is NOT a step in the process of training a neural network?
 Forward Propagation
 Backward Propagation
 Gradient Descent
 Feature Extraction (True)
 Weight Initialization
 Model Evaluation

2. In the context of neural networks, what does the term "batch size" refer to?
 The size of the dataset used for validation.
 The number of epochs in the training process.
 The number of layers in the neural network.
 The number of samples processed in a single forward/backward pass. (True)
 The number of neurons in the output layer.
 The learning rate used for weight updates.

3. What is the purpose of an activation function in a neural network?


 To display visualizations of data.
 To filter out irrelevant information from the input.
 To add extra layers to the neural network.
 To calculate the total number of neurons.
 None of these (True)

4. What is the purpose of the learning rate in gradient descent optimization during neural
network training?
 It determines the number of hidden layers in the network.
 It specifies the size of each training batch.
 It controls the speed at which the model converges to a solution. (True)
 It calculates the number of parameters in the network.
 It defines the activation function used in each layer.
 It sets the number of output neurons in the network.

5. What is the purpose of the "softmax" activation function in the output layer of a neural
network used for multi-class classification?
 It converts all negative values to zero.
 It applies dropout regularization to the output.
 It maps the output to a probability distribution over classes. (True)
 It calculates the mean of the output values.
 It computes the sine of the output values.
 It performs element-wise multiplication of the output.
6. What is the MNIST dataset used for?
 Image classification of animals
 Speech recognition
 Handwritten digit recognition (True)
 Natural language processing
 Object detection
Quiz No: 22
1. What is the architecture of a Vanilla Autoencoder?
 It has only one layer for both encoding and decoding.
 It consists of an encoder and a decoder, both with one layer each. (True)
 It has multiple encoders and a single decoder.
 It uses a recurrent neural network structure.
 It doesn't have a fixed architecture.
 It consists of an encoder and a decoder, both with multiple layers.

2.Which of the following statements is true about Vanilla and Denoising Autoencoders?
 Vanilla Autoencoders always perform better than Denoising Autoencoders.
 Denoising Autoencoders are only used for image data.
 Denoising Autoencoders can learn more meaningful representations than Vanilla
Autoencoders in the presence of noisy input data. (True)
 Vanilla Autoencoders are exclusively used for text data.
 Denoising Autoencoders have fewer layers compared to Vanilla Autoencoders.
 Vanilla Autoencoders are only used for data compression.

3. In a Denoising Autoencoder, what role does the corruption process play during training?
It increases the computational complexity of the model.
 It ensures that the network memorizes the noisy data.
 It makes the training process faster.
 It encourages the model to learn robust features by reducing sensitivity to input
variations. (True)
 It helps the model to overfit the data. It decreases the overall accuracy of the model.

4.How does a Denoising Autoencoder achieve noise removal?


 By training on noisy labels.
 By increasing the learning rate during training.
 By using a larger batch size.
 By corrupting the input data with noise and learning to reconstruct the original data.
(True)
 By applying dropout only during the decoding phase.
 By using a specialized noise removal layer.

5.What is an Autoencoder?
 A machine learning model that classifies data into multiple classes.
 A model used for natural language processing tasks.
 A neural network architecture used for dimensionality reduction and reconstruction of
input data. (True)
 A type of reinforcement learning algorithm.
 An optimization technique used in unsupervised learning.
 A model for time series prediction.
6. What is the bottleneck layer in an Autoencoder?
 The layer responsible for introducing noise during training.
 The output layer of the encoder.
 The input layer of the decoder.
 The layer with the highest number of neurons.
 The layer with the lowest number of neurons. (True)
Quiz No: 23
1. In a Vanilla Autoencoder, which layer is responsible for reconstructing the input data?
 Encoding Layer
 Decoding Layer (True)
 Fully Connected Layer
 Batch Normalization Layer
 Activation Layer
 Pooling Layer

2. Which activation function is commonly used in the encoding and decoding layers of
Autoencoders?
 Sigmoid
 ReLU (Rectified Linear Unit) (True)
 Tanh (Hyperbolic Tangent)
 Leaky ReLU
 Softmax
 Swish

3. What is the main purpose of a Denoising Autoencoder?


 Data compression
 Image segmentation
 Removing noise from data (True)
 Reinforcement learning
 Object detection
 Sentiment analysis

4. What is a CNN primarily used for?


 Text summarization
 Image classification (True)
 Speech recognition
 Stock market prediction
 DNA sequence analysis
 Sentiment analysis

5. Which layer in a CNN is responsible for reducing the spatial dimensions of the input?
 Convolutional Layer
 MaxPooling Layer (True)
 Dropout Layer
 Flatten Layer
 Batch Normalization Layer
 Activation Layer
6. What is the primary objective of a Vanilla Autoencoder?
 Classification
 Feature extraction (True)
 Text generation
 Reinforcement learning
 Speech synthesis
 Anomaly detection

You might also like