Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

AI (Whole)

Download as pdf or txt
Download as pdf or txt
You are on page 1of 96

1/21/24, 12:22 AM Brief History of Generative AI - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Brief History of Generative AI

Quiz review
Started on Sunday, 21 January 2024, 12:16 AM
State Finished
Completed on Sunday, 21 January 2024, 12:22 AM
Time taken 6 mins 1 sec
Marks 9.00/10.00
Grade 90.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Which model demonstrated that using larger architectures can produce better images?

DCGAN
CycleGAN
45804
Transformer
BigGAN
VAE

The correct answer is: BigGAN

Question 2
Correct

Mark 1.00 out of 1.00 45804


In the context of GANs, what is the role of the Discriminator?

To generate data
To transform data
To decode data
To distinguish between real and generated data
To encode data

45804
The correct answer is: To distinguish between real and generated data

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989540&cmid=3675 1/4
1/21/24, 12:22 AM Brief History of Generative AI - Post Quiz: Attempt review

Question 3
Incorrect

Mark 0.00 out of 1.00

Which of the following is NOT a direct application of the Transformer architecture?

Question answering
Image generation 
Text translation
Image recognition
Text summarization

The correct answer is: Image recognition

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which generative model introduced a stochastic layer that models data in a latent space?

BigGAN
CycleGAN
VAE
Transformer
DCGAN

The correct answer is: VAE


45804
Question 5
Correct

Mark 1.00 out of 1.00

Which model is known for its rules for creating stable and effective AI image-makers?

DCGAN
Transformer

45804
VAE
CycleGAN
BigGAN

The correct answer is: DCGAN

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989540&cmid=3675 2/4
1/21/24, 12:22 AM Brief History of Generative AI - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

What mechanism allows the Transformer model to weigh the importance of different words in a sequence?

Recurrent Mechanism
None of the given options
Encoding Mechanism
Self-Attention Mechanism
Decoding Mechanism

The correct answer is: Self-Attention Mechanism

Question 7
Correct

Mark 1.00 out of 1.00


45804
What is the primary advantage of Transformers over RNNs in terms of processing sequences?

Better attention mechanism


Parallel Processing
None of the given options
Faster convergence
More parameters

The correct answer is: Parallel Processing


45804
Question 8
Correct

Mark 1.00 out of 1.00

Which AI model series by OpenAI, based on the Transformer architecture, is known for generating highly coherent content?

CycleGAN
TransformerXL

45804
GPT series
BERT
ResNet

The correct answer is: GPT series

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989540&cmid=3675 3/4
1/21/24, 12:22 AM Brief History of Generative AI - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

What is the main innovation introduced by the "Attention Is All You Need" paper?

Introduction of RNNs
Introduction of VAEs
Introduction of GANs
Introduction of CNNs
Transformer architecture

The correct answer is: Transformer architecture

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which model can transform horse photos into zebra photos without direct comparison?

DCGAN
CycleGAN
BigGAN
VAE
Transformer

The correct answer is: CycleGAN


45804
◄ Additional Reading 3 - Transformers

Jump to...

Fundamentals of ML - Pre Quiz ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989540&cmid=3675 4/4
1/21/24, 12:06 AM Brief History of Generative AI - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Brief History of Generative AI

Quiz review
Started on Saturday, 20 January 2024, 11:56 PM
State Finished
Completed on Sunday, 21 January 2024, 12:06 AM
Time taken 10 mins 24 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What are the two main components of a GAN?

Input and Output


None of the given options
45804
Generator and Discriminator
Forward and Backward
Encoder and Decoder

The correct answer is: Generator and Discriminator

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which of the following research papers is foundational for Variational Autoencoders (VAEs)?

"Auto-Encoding Variational Bayes"


"Deep Residual Learning for Image Recognition"
"Attention is All You Need"
"Generative Adversarial Nets"
"Mastering Chess and Shogi by Self-Play"

45804
The correct answer is: "Auto-Encoding Variational Bayes"

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989352&cmid=3670 1/4
1/21/24, 12:06 AM Brief History of Generative AI - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a direct application of GANs but rather an outcome of its influence?

Image-to-Image translation
Reinforcement learning in game playing
Generating realistic images
Super-resolution
Style transfer

The correct answer is: Reinforcement learning in game playing

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which pioneering research in Generative AI specifically emphasized the generation of text sequences?

"Understanding Machine Learning: From Theory to Algorithms"


"DeepFace: Closing the Gap to Human-Level Performance in Face Recognition"
"A Neural Algorithm of Artistic Style"
"Sequence to Sequence Learning with Neural Networks"
"Visualizing and Understanding Convolutional Networks"

45804
The correct answer is: "Sequence to Sequence Learning with Neural Networks"

Question 5
Correct

Mark 1.00 out of 1.00

Who introduced Generative Adversarial Networks (GANs)?

Yoshua Bengio
Geoffrey Hinton

45804
Ian Goodfellow
Yann LeCun
Andrew Ng

The correct answer is: Ian Goodfellow

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989352&cmid=3670 2/4
1/21/24, 12:06 AM Brief History of Generative AI - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which architecture is primarily associated with attention mechanisms?

GAN
CNN
Transformer
VAE
RNN

The correct answer is: Transformer

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which model uses a probabilistic approach to encode and decode data?

VAE
DCGAN
CycleGAN
Transformer
BigGAN

The correct answer is: VAE


45804
Question 8
Correct

Mark 1.00 out of 1.00

In which year were Generative Adversarial Networks (GANs) first introduced?

2014
2012

45804
2016
2010
2018

The correct answer is: 2014

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989352&cmid=3670 3/4
1/21/24, 12:06 AM Brief History of Generative AI - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

What is the primary purpose of generative models?

Classifying data
None of the given options
Recognizing patterns
Filtering data
Generating new data

The correct answer is: Generating new data

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which model marked a significant milestone in the use of transformers in NLP?

CNN
RNN
BERT
LSTM
GAN

The correct answer is: BERT


45804
◄ Case Study - GEN AI in Fashion

Jump to...

Brief History of Generative AI ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989352&cmid=3670 4/4
1/21/24, 2:18 AM CASE STUDY - GANS - CIFAR - Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Generative Adversarial Networks

Quiz review
Started on Sunday, 21 January 2024, 2:07 AM
State Finished
Completed on Sunday, 21 January 2024, 2:18 AM
Time taken 10 mins 47 secs
Marks 12.00/15.00
Grade 80.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Which challenge refers to the generator producing limited varieties or even the same sample every time?

Convergence Issues
All of the given options
45804
Data Augmentation
Training Instability
Mode Collapse

The correct answer is: Mode Collapse

Question 2
Correct

Mark 1.00 out of 1.00 45804


In the discriminator's code, which layer helps in reducing the dimensions of the input image?

Reshape
BatchNormalization
UpSampling2D
Conv2D with strides
Dense

The correct answer is: Conv2D with strides 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990155&cmid=7551 1/6
1/21/24, 2:18 AM CASE STUDY - GANS - CIFAR - Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which architecture can help address convergence issues in traditional GANs?

DBN
LSTM
WGAN
CNN
RNN

The correct answer is: WGAN

Question 4
Correct

Mark 1.00 out of 1.00


45804
What are the two main components of a GAN?

Generator & Discriminator


Generator & UpSampler
Discriminator & Evaluator
Discriminator & Sampler
Generator & Evaluator

45804
The correct answer is: Generator & Discriminator

Question 5
Correct

Mark 1.00 out of 1.00

How many images are there in each class of the CIFAR-10 dataset?

6000
12000

45804
5000
10000
15000

The correct answer is: 6000

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990155&cmid=7551 2/6
1/21/24, 2:18 AM CASE STUDY - GANS - CIFAR - Quiz: Attempt review

Question 6
Incorrect

Mark 0.00 out of 1.00

Which of the following is NOT a feedback given to the generator during training?

This is a genuine image


This image looks blurry 
This image is pixelated
This image looks like a car
This is a fake image

The correct answer is: This image is pixelated

Question 7
Incorrect

Mark 0.00 out of 1.00


45804
Which technique can help in dealing with training instability in GANs?

Dropout
Data augmentation
Noise addition
All of the given options 
Gradient clipping

The correct answer is: Gradient clipping


45804
Question 8
Correct

Mark 1.00 out of 1.00

In the provided code, why is discriminator.trainable set to False when setting up the combined system?

To prevent overfitting
To speed up training

45804
To increase discriminator's accuracy
To make sure only the generator is trained in this step
None of the given options

The correct answer is: To make sure only the generator is trained in this step

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990155&cmid=7551 3/6
1/21/24, 2:18 AM CASE STUDY - GANS - CIFAR - Quiz: Attempt review

Question 9
Incorrect

Mark 0.00 out of 1.00

What is used to refine the models during training?

Conv2D
All of the given options 
LeakyReLU
Adam Optimizer
Batch Normalization

The correct answer is: Adam Optimizer

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which activation function is used in the final layer of the generator model?

leakyrelu
tanh
sigmoid
relu
softmax

The correct answer is: tanh


45804
Question 11
Correct

Mark 1.00 out of 1.00

Which of the following best describes the role of the generator in a GAN?

To combine images
To produce images

45804
To evaluate the loss
To critique images
None of the given options

The correct answer is: To produce images

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990155&cmid=7551 4/6
1/21/24, 2:18 AM CASE STUDY - GANS - CIFAR - Quiz: Attempt review

Question 12
Correct

Mark 1.00 out of 1.00

What does the discriminator do in a GAN?

Creates images
Evaluates if an image is real or fake
Combines images
Enhances image resolution
Both create and evaluate images

The correct answer is: Evaluates if an image is real or fake

Question 13
Correct

Mark 1.00 out of 1.00


45804
In the generator code, what is the purpose of the Reshape layer?

To critique the images


To reshape the dense layer into a 3D tensor for images
To flatten the images
To normalize the image values
To upsample the images

45804
The correct answer is: To reshape the dense layer into a 3D tensor for images

Question 14
Correct

Mark 1.00 out of 1.00

During training, what does the generator use to improve itself?

Feedback from both the user and the discriminator


Real images

45804
Feedback from the user
CIFAR-10 dataset
Feedback from the discriminator

The correct answer is: Feedback from the discriminator

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990155&cmid=7551 5/6
1/21/24, 2:18 AM CASE STUDY - GANS - CIFAR - Quiz: Attempt review

Question 15
Correct

Mark 1.00 out of 1.00

Why might someone want to use GANs on the CIFAR-10 dataset?

To critique the images in the dataset


To delete images from the dataset
To reduce the size of the dataset
To generate novel and relevant images to augment dataset
To classify the images in the dataset

The correct answer is: To generate novel and relevant images to augment dataset

◄ Case Study - GAN - CIFAR

Jump to... 45804


Sequence Generation with RNNs - Pre Quiz ►

45804

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990155&cmid=7551 6/6
1/22/24, 1:29 AM Case Study - Transformers in Machine Translation - Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Transformers and Attention Mechanisms

Quiz review
Started on Monday, 22 January 2024, 1:25 AM
State Finished
Completed on Monday, 22 January 2024, 1:29 AM
Time taken 4 mins 11 secs
Marks 15.00/15.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What unique aspect is GlobeTech exploring to further enhance translations using Transformers?

45804
Improving voice recognition quality
Reducing translation time further
Enhancing graphics quality
Offering translations considering regional dialects and nuances
Using sentiment analysis on translations

The correct answer is: Offering translations considering regional dialects and nuances

Question 2
Correct

Mark 1.00 out of 1.00 45804


How did Transformers improve GlobeTech's user interface experience for users of different languages?

By offering more payment options


By adding more interactive elements
By providing real-time translations of UI elements
By changing the website layout
By enhancing graphics

45804
The correct answer is: By providing real-time translations of UI elements

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005842&cmid=7550 1/6
1/22/24, 1:29 AM Case Study - Transformers in Machine Translation - Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

What was a major challenge faced by GlobeTech in their previous MT methods?

Graphics
Speed
Real-time Voice Translations
Contextual Translation
Interactivity

The correct answer is: Contextual Translation

Question 4
Correct

Mark 1.00 out of 1.00


45804
How did GlobeTech offer real-time customer support in multiple languages?

Using rule-based translations


By hiring multilingual agents
Integrating Transformer-based MT into their chatbots
Using CNNs
Using Recurrent Networks

45804
The correct answer is: Integrating Transformer-based MT into their chatbots

Question 5
Correct

Mark 1.00 out of 1.00

How did the processing capabilities of Transformers affect GlobeTech's translation time?

Reduced it drastically
Increased server costs

45804
Made it much longer
Had no effect
Made it slightly faster

The correct answer is: Reduced it drastically

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005842&cmid=7550 2/6
1/22/24, 1:29 AM Case Study - Transformers in Machine Translation - Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Traditional MT models required extensive what for each new language?

Debugging
Re-analysis
Refactoring
Re-training and fine-tuning
Re-programming

The correct answer is: Re-training and fine-tuning

Question 7
Correct

Mark 1.00 out of 1.00


45804
The attention mechanism in Transformers allows the model to focus on what?

Different parts of the output sentence


The graphics embedded in the text
Different parts of the input sentence
The beginning of the input sentence
The middle part of the input sentence

45804
The correct answer is: Different parts of the input sentence

Question 8
Correct

Mark 1.00 out of 1.00

Which paper introduced the Transformer architecture?

"Neural Machine Translation"


"Attention Is All You Need"

45804
"Improving Language Understanding by Generative Models"
"Mastering the Game of Go"
"Learning Deep Architectures"

The correct answer is: "Attention Is All You Need"

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005842&cmid=7550 3/6
1/22/24, 1:29 AM Case Study - Transformers in Machine Translation - Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

How did Transformers improve GlobeTech's scalability issue for new languages?

Introduced RNNs
Introduced LSTM
Used Gradient Boosting
Implemented rule-based systems
Leveraged pre-trained models like BERT and GPT

The correct answer is: Leveraged pre-trained models like BERT and GPT

Question 10
Correct

Mark 1.00 out of 1.00


45804
Combining voice recognition and Transformers will help GlobeTech offer what?

Real-time voice translations for customer support


Voice reminders for products
Music recommendations based on voice searches
Voice-activated animations
Voice-activated games

45804
The correct answer is: Real-time voice translations for customer support

Question 11
Correct

Mark 1.00 out of 1.00

Why can we say that Transformers brought a paradigm shift in machine translation?

They changed the way websites were designed


They integrated voice translations into all platforms

45804
They introduced new hardware requirements
They made translations context-aware and faster
They made MT completely manual

The correct answer is: They made translations context-aware and faster

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005842&cmid=7550 4/6
1/22/24, 1:29 AM Case Study - Transformers in Machine Translation - Quiz: Attempt review

Question 12
Correct

Mark 1.00 out of 1.00

After adopting Transformer-based MT, by how much did GlobeTech reduce translation-related complaints?

0.5
0.4
0.2
0.1
0.3

The correct answer is: 0.4

Question 13
Correct

Mark 1.00 out of 1.00


45804
Why did GlobeTech's product descriptions sound off with earlier MT models?

Struggled with contextual meaning, especially with long sentences


They were too short
They lacked graphics
They had many hyperlinks
They lacked interactive elements

45804
The correct answer is: Struggled with contextual meaning, especially with long sentences

Question 14
Correct

Mark 1.00 out of 1.00

What technology does GlobeTech plan to integrate with Transformers for customer support in the future?

Image recognition
Voice recognition

45804
Text summarization
Augmented reality
Gesture recognition

The correct answer is: Voice recognition

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005842&cmid=7550 5/6
1/22/24, 1:29 AM Case Study - Transformers in Machine Translation - Quiz: Attempt review

Question 15
Correct

Mark 1.00 out of 1.00

What unique mechanism in Transformers aids in understanding context?

Backpropagation
Dropout
Self-attention
LSTM cells
CNN layers

The correct answer is: Self-attention

◄ Case Study - Transformers in Machine Translation

Jump to... 45804


Generative AI Applications ►

45804

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005842&cmid=7550 6/6
1/21/24, 1:44 AM CASE STUDY - VAE Application - Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Variational Autoencoders

Quiz review
Started on Sunday, 21 January 2024, 1:34 AM
State Finished
Completed on Sunday, 21 January 2024, 1:44 AM
Time taken 10 mins 23 secs
Marks 14.00/15.00
Grade 93.33 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What type of dataset does the manufacturing plant collect?

Text Dataset
Tabular Dataset
45804
Time Series Dataset
Audio Dataset
Image Dataset

The correct answer is: Time Series Dataset

Question 2
Correct

Mark 1.00 out of 1.00 45804


What two components combine to form the VAE's loss?

L1 loss and L2 loss


MSE and Cross-entropy
Classification error and Regression loss
KL divergence and Cross-entropy
MSE and KL divergence

The correct answer is: MSE and KL divergence 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990039&cmid=7549 1/6
1/21/24, 1:44 AM CASE STUDY - VAE Application - Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

What is a primary application of VAEs mentioned in the case study?

Anomaly Detection
Speech Recognition
Image Classification
Object Detection
Text Summarization

The correct answer is: Anomaly Detection

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which of the following is NOT an attribute in the given data?

Pressure
Temperature
Humidity
Timestamp
Vibration

The correct answer is: Humidity


45804
Question 5
Correct

Mark 1.00 out of 1.00

Which is NOT a challenge in implementing VAEs for this use-case?

Increasing data storage costs


Threshold Setting

45804
Model Training
Data Quality
Latency

The correct answer is: Increasing data storage costs

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990039&cmid=7549 2/6
1/21/24, 1:44 AM CASE STUDY - VAE Application - Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Why is understanding the VAE's outputs challenging?

They are highly interpretable


They are too simplistic
They are always correct
They use an unknown language
They can be complex and non-intuitive

The correct answer is: They can be complex and non-intuitive

Question 7
Correct

Mark 1.00 out of 1.00


45804
Over time, due to certain changes, what might be required of the VAE model?

Disintegration
Manual recalibration
Shrinking
Continuous adaptation
Reformatting

The correct answer is: Continuous adaptation


45804
Question 8
Correct

Mark 1.00 out of 1.00

What does the VAE attempt to minimize during training?

Latent space dimensions


Loss

45804
Data input size
Training time
Validation accuracy

The correct answer is: Loss

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990039&cmid=7549 3/6
1/21/24, 1:44 AM CASE STUDY - VAE Application - Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

What is the y-axis label of the chart visualizing the error?

Latent Space
Data Value
Anomaly Score
Reconstruction Error
Timestamp

The correct answer is: Reconstruction Error

Question 10
Correct

Mark 1.00 out of 1.00


45804
What is the VAE trained to learn effectively?

A visual representation of the data


A highly detailed representation of the data
A textual description of the data
A noisy representation of the data
A compressed representation of the data

45804
The correct answer is: A compressed representation of the data

Question 11
Incorrect

Mark 0.00 out of 1.00

For how many epochs is the VAE trained?

100 
25

45804
40
50
10

The correct answer is: 50

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990039&cmid=7549 4/6
1/21/24, 1:44 AM CASE STUDY - VAE Application - Quiz: Attempt review

Question 12
Correct

Mark 1.00 out of 1.00

What criterion is used to determine if a data point is anomalous?

If its error is above mean error


If its error is above median error
If its error is above the 99th percentile
If its error is below mean error
If its error is in the top 10%

The correct answer is: If its error is above the 99th percentile

Question 13
Correct

Mark 1.00 out of 1.00


45804
Why is data preprocessing required before training the VAE?

To make the data look visually appealing


To ensure it is suitable for training
To make the data larger
To make the data unreadable
To introduce errors into the data

45804
The correct answer is: To ensure it is suitable for training

Question 14
Correct

Mark 1.00 out of 1.00

In the VAE, what does the sampling function introduce?

Parallelism
Determinism

45804
Recursion
Linearity
Randomness

The correct answer is: Randomness

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990039&cmid=7549 5/6
1/21/24, 1:44 AM CASE STUDY - VAE Application - Quiz: Attempt review

Question 15
Correct

Mark 1.00 out of 1.00

How is the data divided for training the VAE?

60-40
90-10
70-30
80-20
50-50

The correct answer is: 80-20

◄ Case Study - Variational Auto Encoder

Jump to... 45804


Generative Adversarial Networks - Pre Quiz ►

45804

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990039&cmid=7549 6/6
1/21/24, 12:47 AM Fundamentals of ML - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Fundamentals of Machine Learning and Neural Networks

Quiz review
Started on Sunday, 21 January 2024, 12:42 AM
State Finished
Completed on Sunday, 21 January 2024, 12:47 AM
Time taken 4 mins 39 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Which function introduces non-linearity in a neural network?

Linear Function
Bias Function
45804
Weight Function
Activation Function
Loss Function

The correct answer is: Activation Function

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which application of ML is used to detect unusual patterns in data?

Clustering
Ranking
Regression
Classification
Anomaly Detection

The correct answer is: Anomaly Detection 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989761&cmid=3677 1/4
1/21/24, 12:47 AM Fundamentals of ML - Post Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which of the following is a common activation function in neural networks?

Polynomial Function
Weighted Sum
Bias Activation
Linear Function
ReLU (Rectified Linear Unit)

The correct answer is: ReLU (Rectified Linear Unit)

Question 4
Correct

Mark 1.00 out of 1.00


45804
What is the primary purpose of backpropagation?

Data preprocessing
Forward propagation of data
Activation of neurons
Adjusting weights based on the error
Initialization of weights

45804
The correct answer is: Adjusting weights based on the error

Question 5
Correct

Mark 1.00 out of 1.00

In which type of ML does an agent learn by interacting with an environment?

Regression
Unsupervised Learning

45804
Clustering
Supervised Learning
Reinforcement Learning

The correct answer is: Reinforcement Learning

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989761&cmid=3677 2/4
1/21/24, 12:47 AM Fundamentals of ML - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

In a neural network, what does a neuron compute?

A fixed value
The error of the network
The learning rate
A weighted sum followed by an activation function
The gradient of the loss

The correct answer is: A weighted sum followed by an activation function

Question 7
Correct

Mark 1.00 out of 1.00


45804
What is the role of the loss function in training a neural network?

To initialize the weights


To define the network architecture
To quantify the difference between predicted and actual values
To introduce non-linearity
To activate the neurons

45804
The correct answer is: To quantify the difference between predicted and actual values

Question 8
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a layer type in a typical neural network?

Quantum Layer
Output Layer

45804
Hidden Layer
Convolutional Layer
Input Layer

The correct answer is: Quantum Layer

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989761&cmid=3677 3/4
1/21/24, 12:47 AM Fundamentals of ML - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

How is a neural network's performance typically evaluated during training?

Using a validation set


Using the test data
Using the activation functions
Using the training data
Using the weights

The correct answer is: Using a validation set

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which of the following is a challenge in training deep neural networks?

Vanishing/Exploding gradients
Too few neurons
Small datasets
Linear activation functions
All neurons activating at once

45804
The correct answer is: Vanishing/Exploding gradients

◄ Fundamentals of Machine Learning and Neural Networks

Jump to...

Introduction to Generative Models - Pre Quiz ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989761&cmid=3677 4/4
1/21/24, 12:26 AM Fundamentals of ML - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Fundamentals of Machine Learning and Neural Networks

Quiz review
Started on Sunday, 21 January 2024, 12:22 AM
State Finished
Completed on Sunday, 21 January 2024, 12:26 AM
Time taken 4 mins 18 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Which activation function outputs a value between 0 and 1?

Leaky ReLU
Linear
45804
Sigmoid
Hyperbolic Tangent (tanh)
Rectified Linear Unit (ReLU)

The correct answer is: Sigmoid

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which application of ML is used to group similar items?

Regression
Clustering
Ranking
Classification
Recommendation

The correct answer is: Clustering 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989595&cmid=3676 1/4
1/21/24, 12:26 AM Fundamentals of ML - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

What is the main difference between regression and classification?

Both are the same


Regression uses labeled data, Classification doesn't
Regression is unsupervised
Classification is unsupervised
Regression predicts a continuous output, Classification predicts a discrete label

The correct answer is: Regression predicts a continuous output, Classification predicts a discrete label

Question 4
Correct

Mark 1.00 out of 1.00


45804
What is the primary purpose of a loss function in training neural networks?

To initialize weights
To define the network's architecture
To speed up training
To activate neurons
To quantify the difference between predicted and actual values

45804
The correct answer is: To quantify the difference between predicted and actual values

Question 5
Correct

Mark 1.00 out of 1.00

What is the primary goal of machine learning?

To design new algorithms


To allow computers to learn from data

45804
None of the given options
To program explicit rules for a task
To increase computational speed

The correct answer is: To allow computers to learn from data

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989595&cmid=3676 2/4
1/21/24, 12:26 AM Fundamentals of ML - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which component of a neural network is responsible for combining inputs and passing them to the next layer?

Bias
Weight
Activation Function
Neuron (or Node)
Layer

The correct answer is: Neuron (or Node)

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which of the following is NOT a type of machine learning?

Unsupervised Learning
Recursive Learning
Reinforcement Learning
Semi-supervised Learning
Supervised Learning

The correct answer is: Recursive Learning


45804
Question 8
Correct

Mark 1.00 out of 1.00

In the context of neural networks, what does the term "backpropagation" refer to?

The forward flow of data


The activation of neurons in the hidden layer

45804
The method of adjusting weights based on the error
The initial random assignment of weights
The process of adding more layers

The correct answer is: The method of adjusting weights based on the error

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989595&cmid=3676 3/4
1/21/24, 12:26 AM Fundamentals of ML - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a common machine learning algorithm?

Quantum Entanglement
Neural Networks
K-Means Clustering
Support Vector Machines
Decision Trees

The correct answer is: Quantum Entanglement

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which of the following is a technique to prevent overfitting in neural networks?

Gradient Clipping
Increasing the number of layers
Dropout
Using a larger dataset
Learning Rate Adjustment

The correct answer is: Dropout


45804
◄ Brief History of Generative AI - Post Quiz

Jump to...

Fundamentals of Machine Learning and Neural Networks ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989595&cmid=3676 4/4
1/21/24, 2:07 AM Generative Adversarial Networks - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Generative Adversarial Networks

Quiz review
Started on Sunday, 21 January 2024, 2:00 AM
State Finished
Completed on Sunday, 21 January 2024, 2:07 AM
Time taken 6 mins 46 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Which GAN variant can be conditioned on labels to generate specific outputs?

Minimax GAN
None of the given options
45804
Conditional GAN
Progressive GAN
Mode Collapse GAN

The correct answer is: Conditional GAN

Question 2
Correct

Mark 1.00 out of 1.00 45804


The generator's objective in GANs is to...

None of the given options


Reduce mode collapse
Fool the discriminator
Improve model accuracy
Classify real vs. fake

The correct answer is: Fool the discriminator 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990130&cmid=3683 1/4
1/21/24, 2:07 AM Generative Adversarial Networks - Post Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Mode collapse is problematic because...

It limits the diversity of generated outputs


It requires more data
It speeds up training
It makes the discriminator weak
None of the given options

The correct answer is: It limits the diversity of generated outputs

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which is NOT a real-world application of GANs?

Real-time weather prediction


Style transfer
Super-resolution imaging
Data augmentation
Art generation

45804
The correct answer is: Real-time weather prediction

Question 5
Correct

Mark 1.00 out of 1.00

In GANs, if the discriminator becomes too powerful, what can happen?

The training process speeds up


None of the given options

45804
The model achieves perfect accuracy
The generator may struggle to improve
The generator becomes powerful too

The correct answer is: The generator may struggle to improve

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990130&cmid=3683 2/4
1/21/24, 2:07 AM Generative Adversarial Networks - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which statement about GANs is true?

They are a type of supervised learning


They only work with images
None of the given options
They always converge to a solution
They can generate new, previously unseen data

The correct answer is: They can generate new, previously unseen data

Question 7
Correct

Mark 1.00 out of 1.00


45804
In the minimax game of GANs, what is the discriminator's goal?

Minimize its own loss


None of the given options
Generate realistic data
Distinguish between real and fake data
Maximize the generator's loss

45804
The correct answer is: Distinguish between real and fake data

Question 8
Correct

Mark 1.00 out of 1.00

What is a challenge in evaluating the performance of GANs?

They require large datasets


They always outperform other models

45804
None of the given options
They are too fast
Determining the quality of generated data

The correct answer is: Determining the quality of generated data

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990130&cmid=3683 3/4
1/21/24, 2:07 AM Generative Adversarial Networks - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which GAN variant focuses on gradually increasing the resolution of generated images?

Progressive GAN
Minimax GAN
Mode Collapse GAN
None of the given options
Conditional GAN

The correct answer is: Progressive GAN

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which component of a GAN tries to produce fake data?

None of the given options


Discriminator
Decoder
Generator
Encoder

The correct answer is: Generator


45804
◄ Generative Adversarial Networks

Jump to...

Case Study - GAN - CIFAR ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990130&cmid=3683 4/4
1/21/24, 1:50 AM Generative Adversarial Networks - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Generative Adversarial Networks

Quiz review
Started on Sunday, 21 January 2024, 1:45 AM
State Finished
Completed on Sunday, 21 January 2024, 1:50 AM
Time taken 5 mins 37 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What does GAN stand for?

None of the given options 45804


Generalized Artificial Network
Generative Analytical Network
Generative Adversarial Network
Gradient Augmented Network

The correct answer is: Generative Adversarial Network

Question 2
Correct

Mark 1.00 out of 1.00 45804


In GANs, which component is responsible for evaluating the authenticity of data?

Encoder
None of the given options
Discriminator
Generator
Decoder

The correct answer is: Discriminator 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990079&cmid=3682 1/4
1/21/24, 1:50 AM Generative Adversarial Networks - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which component of a GAN is responsible for generating new data samples?

Optimizer
Decoder
Discriminator
Encoder
Generator

The correct answer is: Generator

Question 4
Correct

Mark 1.00 out of 1.00


45804
What is a challenge faced during GAN training due to the minimax game concept?

Oscillations and non-convergence


Discriminator becoming too weak
Generator producing only a single mode
Quick convergence to a suboptimal solution
Overfitting to the training data

45804
The correct answer is: Oscillations and non-convergence

Question 5
Correct

Mark 1.00 out of 1.00

Progressive GANs are designed to address which challenge in traditional GANs?

Inability to generate colored images


Slow training speeds

45804
Mode collapse
Discriminator overpowering the generator
Training stability and generating high-resolution images

The correct answer is: Training stability and generating high-resolution images

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990079&cmid=3682 2/4
1/21/24, 1:50 AM Generative Adversarial Networks - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

The training process of GANs is often likened to which game?

Poker
Chess
None of the given options
Minimax
Sudoku

The correct answer is: Minimax

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which type of GAN allows for generating data based on specific categories?

None of the given options


Minimax GAN
Mode Collapse GAN
Progressive GAN
Conditional GAN

The correct answer is: Conditional GAN


45804
Question 8
Correct

Mark 1.00 out of 1.00

Which of the following is a real-world application where GANs have shown significant promise?

Image-to-image translation
Image classification

45804
Text summarization
Time series forecasting
Speech recognition

The correct answer is: Image-to-image translation

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990079&cmid=3682 3/4
1/21/24, 1:50 AM Generative Adversarial Networks - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

What is mode collapse in the context of GANs?

When the model underfits


When the model overfits
When the discriminator becomes too powerful
When the generator produces limited varieties of outputs
When the model converges too quickly

The correct answer is: When the generator produces limited varieties of outputs

Question 10
Correct

Mark 1.00 out of 1.00


45804
In the GAN architecture, what is the primary goal of the Discriminator?

Minimize the loss function


Distinguish between real and generated samples
Ensure mode diversity
Generate realistic data samples
Replicate the generator's output

45804
The correct answer is: Distinguish between real and generated samples

◄ CASE STUDY - VAE Application - Quiz

Jump to...

Generative Adversarial Networks ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990079&cmid=3682 4/4
1/22/24, 1:47 AM Generative AI Applications in Key Industries: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Generative AI in Industry and Real-World Applications

Quiz review
Started on Monday, 22 January 2024, 1:40 AM
State Finished
Completed on Monday, 22 January 2024, 1:47 AM
Time taken 6 mins 16 secs
Marks 9.00/10.00
Grade 90.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

In the realm of manufacturing, how does generative AI impact the design process?

By enhancing MRI visuals


By crafting ad content
45804
By facilitating binary decisions
By creating product designs
By monitoring crop health

The correct answer is: By creating product designs

Question 2
Incorrect

Mark 0.00 out of 1.00 45804


What differentiates Google Bard's data access from ChatGPT?

Bard is built on GPT-4


Bard extracts real-time information
ChatGPT offers improved visuals 
ChatGPT employs discriminative AI
Bard has more visual capabilities

45804
The correct answer is: Bard extracts real-time information

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005931&cmid=3796 1/4
1/22/24, 1:47 AM Generative AI Applications in Key Industries: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

During the harvesting phase, how does AI offer a boon to the agricultural sector?

By forming individual educational pathways


By distinguishing inferior plants
By translating marketing content
By enhancing financial processes
By amplifying equipment resilience

The correct answer is: By distinguishing inferior plants

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which industry utilizes AI for personalized care programs enhancing patient recovery?

Cybersecurity
Manufacturing
Healthcare
Education
Advertising

The correct answer is: Healthcare


45804
Question 5
Correct

Mark 1.00 out of 1.00

DALL-E's image generation can be optimal for which of the following applications?

Translating ad content
Designing book covers

45804
Simulating cyber risk scenarios
Binary choice models
Enhancing banking interactions

The correct answer is: Designing book covers

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005931&cmid=3796 2/4
1/22/24, 1:47 AM Generative AI Applications in Key Industries: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which AI methodology specializes in data set differentiation?

Binary AI
Discriminative AI
Generative AI
Transformer AI
Visual AI

The correct answer is: Discriminative AI

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which conversational AI is not constructed on the Transformer neural network foundation?

DALL-E
LaMDA
Bing AI
ChatGPT
Google Bard

The correct answer is: DALL-E


45804
Question 8
Correct

Mark 1.00 out of 1.00

For which sector does generative AI replicate potential threat environments to bolster proactive defense?

Advertising
Finance

45804
Agriculture
Cybersecurity
Education

The correct answer is: Cybersecurity

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005931&cmid=3796 3/4
1/22/24, 1:47 AM Generative AI Applications in Key Industries: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which AI model, developed by Google, is designed to engage in open-ended conversations, often generating creative responses to user prompts?

Bing AI
DALL-E
Google Bard
ChatGPT
LaMDA

The correct answer is: LaMDA

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which AI platform, integrated into Microsoft's Bing, delivers instant query answers?

LaMDA
DALL-E
Bing AI
ChatGPT
Google Bard

The correct answer is: Bing AI


45804
◄ Case Study - Generative AI Applications in Key Industries

Jump to...

Case Study - Generative AI Tools ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005931&cmid=3796 4/4
1/22/24, 1:52 AM Generative AI Tools: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Generative AI in Industry and Real-World Applications

Quiz review
Started on Monday, 22 January 2024, 1:49 AM
State Finished
Completed on Monday, 22 January 2024, 1:52 AM
Time taken 3 mins 11 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Approximately what percentage of false positive rate does AlphaCode by DeepMind have?

0.01
0.02
45804
0.04
0.05
0.03

The correct answer is: 0.04

Question 2
Correct

Mark 1.00 out of 1.00 45804


What is a unique feature of ChatGPT that distinguishes it from Bard by Google?

Built on LaMDA transformer model


No conversational history feature
Ethically developed
Retains conversation history
Designed for human dialogue

The correct answer is: Retains conversation history 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005978&cmid=3797 1/4
1/22/24, 1:52 AM Generative AI Tools: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Bard by Google has limitations in which of the following aspects?

Ethically developed
Programming and software development capabilities
Constantly updated with web information
Limited to English language
Transformer model

The correct answer is: Limited to English language

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which database does GitHub Copilot ground its data on?

Anthropic's Claude database


GPT-4 database
OpenAI Codex and GitHub
DeepMind's AlphaCode repository
Synthesia's scripted prompts

45804
The correct answer is: OpenAI Codex and GitHub

Question 5
Correct

Mark 1.00 out of 1.00

Which of the following is NOT an attribute of GPT-4 by OpenAI?

Persistent bias issues


Audio outputs

45804
Large multimodal model
Image and text input
Enhanced creativity and accuracy

The correct answer is: Audio outputs

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005978&cmid=3797 2/4
1/22/24, 1:52 AM Generative AI Tools: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Compared to its predecessor, DALL-E, what is an improved feature of DALL-E 2?

Ethical development
Higher resolution
Requires less purchase credits
Less safety protocols
Same image resolution

The correct answer is: Higher resolution

Question 7
Correct

Mark 1.00 out of 1.00


45804
What is a potential concern when using Code Whisperer by AWS with open-source projects?

It aligns with best practices


Challenges with complex tasks
AWS optimization
It boosts productivity with instant suggestions
Potential open-source legal issues

45804
The correct answer is: Potential open-source legal issues

Question 8
Correct

Mark 1.00 out of 1.00

Cohere Generate primarily targets which type of content?

Quick code generation via language prompts


Conversational tone with Slack integration

45804
Marketing and sales content
Language inputs for image outputs
Video creation from scripted prompts

The correct answer is: Marketing and sales content

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005978&cmid=3797 3/4
1/22/24, 1:52 AM Generative AI Tools: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

How does Claude by Anthropic enhance its safety features?

Ethically developed
By accessing the web
Using "red team" prompts for safety
Slack integration
Emphasizing creativity

The correct answer is: Using "red team" prompts for safety

Question 10
Correct

Mark 1.00 out of 1.00


45804
For which feature might users of the basic plans of Synthesia encounter quality concerns?

Audio quality
Scripted prompts
Efficient basic content generation
Video resolution
Language integrations

The correct answer is: Audio quality


45804
◄ Case Study - Generative AI Tools

Jump to...

Generative AI Influence on Specific Industries: An In-depth Analysis ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005978&cmid=3797 4/4
1/20/24, 11:55 PM Introduction to Generative AI - Post quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Introduction to Generative AI

Quiz review
Started on Saturday, 20 January 2024, 11:50 PM
State Finished
Completed on Saturday, 20 January 2024, 11:55 PM
Time taken 4 mins 57 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

In the context of AI, which model type is more concerned with the underlying distribution of data?

Hybrid AI
Reinforcement AI
45804
Classification AI
Generative AI
Regression AI

The correct answer is: Generative AI

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which AI type is best for predicting outcomes?

Semi-supervised AI
Generative AI
Reinforcement AI
Regression AI
Classification AI

The correct answer is: Regression AI 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989290&cmid=3667 1/4
1/20/24, 11:55 PM Introduction to Generative AI - Post quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

If an AI system is designed to label images of cats and dogs, it is primarily a _______ model.

Hybrid
Generative
Reinforcement
Unsupervised
Discriminative

The correct answer is: Discriminative

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which is NOT a real-world application of Generative AI?

Producing synthetic voices


Deepfake videos
Generating game environments
Creating virtual fashion designs
Predicting stock market prices

45804
The correct answer is: Predicting stock market prices

Question 5
Correct

Mark 1.00 out of 1.00

What is Generative AI primarily used for?

Regression
Optimization

45804
Generating new data
Data labeling
Classification

The correct answer is: Generating new data

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989290&cmid=3667 2/4
1/20/24, 11:55 PM Introduction to Generative AI - Post quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

How does Generative AI differ from Classification AI?

It's more accurate


It requires more data
It's easier to implement
It generates new data rather than categorizing existing data
It's faster

The correct answer is: It generates new data rather than categorizing existing data

Question 7
Correct

Mark 1.00 out of 1.00


45804
Generative AI can be used to create which of the following?

New artworks and music pieces


Regression models
Data labels
Classification categories
Decision boundaries

45804
The correct answer is: New artworks and music pieces

Question 8
Correct

Mark 1.00 out of 1.00

Which statement best describes the role of Generative AI?

It focuses on generating data based on learned patterns


It is exclusively used in robotics

45804
It is primarily used for data sorting
It is best suited for regression tasks
It is the oldest form of AI

The correct answer is: It focuses on generating data based on learned patterns

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989290&cmid=3667 3/4
1/20/24, 11:55 PM Introduction to Generative AI - Post quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which of the following is a direct application of Generative AI in the entertainment industry?

Predicting movie success


Translating movie scripts
Automating video editing
Creating realistic CGI characters
Recommending movies to users

The correct answer is: Creating realistic CGI characters

Question 10
Correct

Mark 1.00 out of 1.00


45804
Why is Generative AI considered significant in the realm of artificial intelligence?

It exclusively works with images


It speeds up training processes
It reduces the need for large datasets
It can produce new, previously unseen data samples
It simplifies complex algorithms

45804
The correct answer is: It can produce new, previously unseen data samples

◄ Introduction to Generative AI

Jump to...

Case Study - GEN AI in Fashion ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989290&cmid=3667 4/4
1/20/24, 11:33 PM Introduction to Generative AI - Pre quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Introduction to Generative AI

Quiz review
Started on Saturday, 20 January 2024, 11:28 PM
State Finished
Completed on Saturday, 20 January 2024, 11:33 PM
Time taken 5 mins 34 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What does AI stand for?

Application Interface
Automated Information
45804
Automated Interaction
Artificial Intelligence
Advanced Integration

The correct answer is: Artificial Intelligence

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which of the following is NOT a type of AI?

Unsupervised AI
Supervised AI
Generative AI
Reinforcement AI
Generative Art

The correct answer is: Generative Art 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989037&cmid=3666 1/4
1/20/24, 11:33 PM Introduction to Generative AI - Pre quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which type of AI is primarily concerned with how data is generated rather than how it's separated?

Reinforcement Learning
Discriminative AI
Supervised Learning
Generative AI
Unsupervised Learning

The correct answer is: Generative AI

Question 4
Correct

Mark 1.00 out of 1.00


45804
In which application is Generative AI NOT typically used?

Creating art
Designing virtual environments
Automating customer service chats
Generating music
Producing realistic video game characters

45804
The correct answer is: Automating customer service chats

Question 5
Correct

Mark 1.00 out of 1.00

Which AI type primarily focuses on labeling data?

Semi-supervised AI
Generative AI

45804
Reinforcement AI
Supervised AI
Regression AI

The correct answer is: Supervised AI

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989037&cmid=3666 2/4
1/20/24, 11:33 PM Introduction to Generative AI - Pre quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which statement best defines Generative AI?

AI that automates repetitive tasks


AI that understands human emotions
AI that predicts future trends
AI that classifies data
AI that can generate new data samples

The correct answer is: AI that can generate new data samples

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which of the following is a real-world example of Generative AI?

Generating realistic human faces in movies


Predicting stock market prices
Sorting emails
Translating languages
Automating cars

45804
The correct answer is: Generating realistic human faces in movies

Question 8
Correct

Mark 1.00 out of 1.00

Which of the following fields can utilize Generative AI to create new, original content or simulations?

Data Analysis
Art and Music

45804
Transportation
Banking
E-commerce

The correct answer is: Art and Music

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989037&cmid=3666 3/4
1/20/24, 11:33 PM Introduction to Generative AI - Pre quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Generative AI is closely related to which type of models?

Generative models
Clustering models
Regression models
Decision trees
Classification models

The correct answer is: Generative models

Question 10
Correct

Mark 1.00 out of 1.00


45804
What distinguishes Generative AI from Discriminative AI?

Generative is for images, Discriminative for text


Generative is older, Discriminative is newer
Generative models data distribution, while Discriminative models the boundary between classes
Generative focuses on labeling, Discriminative on generating
Both are the same

45804
The correct answer is: Generative models data distribution, while Discriminative models the boundary between classes

Jump to...

Introduction to Generative AI ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989037&cmid=3666 4/4
1/21/24, 1:08 AM Introduction to Generative Models - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Introduction to Generative Models

Quiz review
Started on Sunday, 21 January 2024, 1:03 AM
State Finished
Completed on Sunday, 21 January 2024, 1:08 AM
Time taken 5 mins 37 secs
Marks 9.00/10.00
Grade 90.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What does a probability distribution provide?

45804
A method for generating new data
A training method for models
A mathematical description of outcomes for a random variable
A decision boundary for classification
A measure of model error

The correct answer is: A mathematical description of outcomes for a random variable

Question 2
Correct

Mark 1.00 out of 1.00 45804


Within generative models, what function does the discriminator serve in GANs?

To optimize the generator


To distinguish between real and generated data
To generate new data
To capture the joint probability
To calculate the likelihood

45804
The correct answer is: To distinguish between real and generated data

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989909&cmid=3679 1/4
1/21/24, 1:08 AM Introduction to Generative Models - Post Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

For what tasks can generative models be applied?

Data generation, denoising, inpainting, and more


Only data generation
Data labeling only
Only denoising
Classification only

The correct answer is: Data generation, denoising, inpainting, and more

Question 4
Correct

Mark 1.00 out of 1.00


45804
How is the likelihood of data given a model symbolized?

P(data | model)
P(model)
P(data & model)
P(model | data)
P(data)

The correct answer is: P(data | model)


45804
Question 5
Correct

Mark 1.00 out of 1.00

Which claim regarding generative models isn't true?

They can generate new data samples


They can be combined with discriminative models for certain tasks

45804
They always require labeled data for training
They can be used in unsupervised learning scenarios
They capture the data distribution

The correct answer is: They always require labeled data for training

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989909&cmid=3679 2/4
1/21/24, 1:08 AM Introduction to Generative Models - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which of the following best describes the difference between generative and discriminative models?

Generative models are always better


Generative models are older in concept
Generative models learn the data distribution, while discriminative models learn the decision boundary
Generative models are used for classification only
Discriminative models can't generate data

The correct answer is: Generative models learn the data distribution, while discriminative models learn the decision boundary

Question 7
Incorrect

Mark 0.00 out of 1.00


45804
Which of the following is NOT a property of likelihood?

It can be used to compare different models


It is a function of model parameters
It is always a probability between 0 and 1 
It is not normalized like a probability
It measures how well a model explains data

45804
The correct answer is: It is not normalized like a probability

Question 8
Correct

Mark 1.00 out of 1.00

Within the architecture of Generative Adversarial Networks (GANs), which duo of fundamental elements are paramount?

Classifier and Regressor


Encoder and Decoder

45804
Activator and Deactivator
Forward and Backward Propagators
Generator and Discriminator

The correct answer is: Generator and Discriminator

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989909&cmid=3679 3/4
1/21/24, 1:08 AM Introduction to Generative Models - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

What's a significant hurdle when training GANs?

Overfitting to the training data


Slow convergence rate
Inability to generate high-resolution images
The discriminator becoming too weak
Mode collapse

The correct answer is: Mode collapse

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which model type aims to capture the joint probability P(x, y)?

Discriminative Model
regression model
Probability Distribution
Both Generative Model and Discriminative Model
Generative Model

The correct answer is: Generative Model


45804
◄ Introduction to Generative Models

Jump to...

Variational Autoencoders - Pre Quiz ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989909&cmid=3679 4/4
1/21/24, 12:51 AM Introduction to Generative Models - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Introduction to Generative Models

Quiz review
Started on Sunday, 21 January 2024, 12:47 AM
State Finished
Completed on Sunday, 21 January 2024, 12:51 AM
Time taken 4 mins 26 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

If a model is better at distinguishing between classes rather than generating data, it is likely a _______.

Generative model
Joint probability model
45804
Discriminative model
Bayesian model
Likelihood model

The correct answer is: Discriminative model

Question 2
Correct

Mark 1.00 out of 1.00 45804


In the context of generative models, what does P(x) represent?

The probability distribution of the data x


The conditional probability of x given y
The posterior probability of x
The likelihood of x
The joint probability of x and y

45804
The correct answer is: The probability distribution of the data x

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989799&cmid=3678 1/4
1/21/24, 12:51 AM Introduction to Generative Models - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a generative model?

Restricted Boltzmann Machines


Variational Autoencoders
Gaussian Mixture Models
Generative Adversarial Networks
Support Vector Machines

The correct answer is: Support Vector Machines

Question 4
Correct

Mark 1.00 out of 1.00


45804
Generative models are primarily used for which of the following tasks?

Classification
Regression
Reinforcement learning
Generating new data samples similar to the input data
Clustering

45804
The correct answer is: Generating new data samples similar to the input data

Question 5
Correct

Mark 1.00 out of 1.00

What is the primary goal of generative models in AI?

To reduce computational cost


To optimize algorithms

45804
To generate new data samples
To analyze data distributions
To classify data

The correct answer is: To generate new data samples

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989799&cmid=3678 2/4
1/21/24, 12:51 AM Introduction to Generative Models - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

In the context of models, what does P(x | y) typically represent?

The distribution of y
The likelihood of y
The generative capacity of x
The probability of x given y
The probability of y given x

The correct answer is: The probability of x given y

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which statement best differentiates generative from discriminative models?

Both models serve the same purpose


Generative models are newer than discriminative models
Generative models are only for images, discriminative for text
Generative models cannot be trained with labeled data
Generative models learn the joint probability distribution, while discriminative models learn the conditional probability

45804
The correct answer is: Generative models learn the joint probability distribution, while discriminative models learn the conditional probability

Question 8
Correct

Mark 1.00 out of 1.00

Which model type is primarily concerned with determining P(y | x)?

Bayesian model
Probability Distribution

45804
Generative Model
Discriminative Model
Both Generative and Discriminative

The correct answer is: Discriminative Model

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989799&cmid=3678 3/4
1/21/24, 12:51 AM Introduction to Generative Models - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which of the following is crucial for understanding the behavior of generative models?

Convolutional layers
Probability distributions and likelihood
Gradient descent
Activation functions
Backpropagation

The correct answer is: Probability distributions and likelihood

Question 10
Correct

Mark 1.00 out of 1.00


45804
What does likelihood measure in the context of a model?

The complexity of the model


The probability of the model being correct
The error rate of the model
How well the model explains the observed data
The generative capacity of the model

45804
The correct answer is: How well the model explains the observed data

◄ Fundamentals of ML - Post Quiz

Jump to...

Introduction to Generative Models ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989799&cmid=3678 4/4
1/22/24, 12:44 AM Sentiment Analysis with RNNs - Case study: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Sequence Generation with RNNs

Quiz review
Started on Monday, 22 January 2024, 12:38 AM
State Finished
Completed on Monday, 22 January 2024, 12:44 AM
Time taken 6 mins 2 secs
Marks 14.00/15.00
Grade 93.33 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What is the role of the `<OOV>` token?

45804
Placeholder for out-of-vocabulary words
Placeholder for numbers
Regular expression matcher
Ignore out-of-vocabulary words
Delete out-of-vocabulary words

The correct answer is: Placeholder for out-of-vocabulary words

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which layer in the RNN model represents words as detailed feature lists?

Embedding Layer
Dropout Layer
Dense Layer
SimpleRNN Layer
LSTM Layer

The correct answer is: Embedding Layer 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005439&cmid=7547 1/6
1/22/24, 12:44 AM Sentiment Analysis with RNNs - Case study: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Why is padding used in the preprocessing step?

For beautification
To reduce memory usage
To increase vocabulary size
To improve accuracy
To handle variable review length

The correct answer is: To handle variable review length

Question 4
Correct

Mark 1.00 out of 1.00


45804
What advantage does LSTM have over traditional RNNs?

Lower memory usage


Requires fewer layers
Simpler architecture
Faster convergence
Tackles the vanishing gradient problem

45804
The correct answer is: Tackles the vanishing gradient problem

Question 5
Correct

Mark 1.00 out of 1.00

What is the purpose of the Dropout layer in the LSTM with Dropout model?

Embedding
Activation function

45804
Regularization to prevent overfitting
Tokenization
Recurrence

The correct answer is: Regularization to prevent overfitting

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005439&cmid=7547 2/6
1/22/24, 12:44 AM Sentiment Analysis with RNNs - Case study: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

What might be a concern if the training accuracy is high but validation accuracy is significantly low?

Model is perfectly trained


Model is overfitting
Model needs more layers
Data is incorrectly labeled
Model is underfitting

The correct answer is: Model is overfitting

Question 7
Correct

Mark 1.00 out of 1.00


45804
In which scenario might you prefer a simple RNN over an LSTM?

Long-range dependencies in data


When high accuracy is a must
Large datasets
Complex sentence structures
Fast training with limited resources

45804
The correct answer is: Fast training with limited resources

Question 8
Correct

Mark 1.00 out of 1.00

Which parameter in `model.fit()` signifies the number of times the model is exposed to the dataset?

epochs
loss

45804
optimizer
validation_data
batch_size

The correct answer is: epochs

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005439&cmid=7547 3/6
1/22/24, 12:44 AM Sentiment Analysis with RNNs - Case study: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Why is the loss function important during model compilation?

Specifies how errors are measured


Specifies number of epochs
Assigns weights to layers
Determines model layers
Adjusts learning rate

The correct answer is: Specifies how errors are measured

Question 10
Correct

Mark 1.00 out of 1.00


45804
How does the model handle reviews of varying lengths?

Uses multiple RNN layers


Uses padding
Uses LSTM layers
Changes tokenizer's vocabulary
Ignores reviews outside a certain length range

The correct answer is: Uses padding


45804
Question 11
Correct

Mark 1.00 out of 1.00

Why might the vanishing gradient problem be a challenge in RNNs?

Reduces training speed


Requires more memory

45804
Makes model evaluation faster
Impedes learning of long-range dependencies
Increases accuracy

The correct answer is: Impedes learning of long-range dependencies

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005439&cmid=7547 4/6
1/22/24, 12:44 AM Sentiment Analysis with RNNs - Case study: Attempt review

Question 12
Correct

Mark 1.00 out of 1.00

In the given LSTM model, which layer(s) help in retaining memory and context?

Dropout layer
Dense layer
SimpleRNN layer
Embedding layer
LSTM layer

The correct answer is: LSTM layer

Question 13
Correct

Mark 1.00 out of 1.00


45804
When using a tokenizer with a fixed number of words, what could be a potential drawback?

Limited understanding due to missed words


Increases memory usage
Slows down training
Enhances accuracy
Simplifies the model

45804
The correct answer is: Limited understanding due to missed words

Question 14
Correct

Mark 1.00 out of 1.00

What is the primary function of an Embedding Layer?

Tokenization
Representing words in dense vector format

45804
Regularization
Handling out-of-vocabulary words
Reducing sequence length

The correct answer is: Representing words in dense vector format

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005439&cmid=7547 5/6
1/22/24, 12:44 AM Sentiment Analysis with RNNs - Case study: Attempt review

Question 15
Incorrect

Mark 0.00 out of 1.00

After training, what can be inferred if the validation loss keeps decreasing but training loss remains high?

Training data is corrupted


Model architecture is flawed
Model is overfitting 
Model is underfitting
Model is perfectly trained

The correct answer is: Model is underfitting

◄ Case Study - Sentiment Analysis with RNNs

Jump to... 45804


Transformers and Attention Mechanisms - Pre Quiz ►

45804

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005439&cmid=7547 6/6
1/22/24, 12:38 AM Sequence Generation with RNNs - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Sequence Generation with RNNs

Quiz review
Started on Monday, 22 January 2024, 12:30 AM
State Finished
Completed on Monday, 22 January 2024, 12:37 AM
Time taken 7 mins 45 secs
Marks 9.00/10.00
Grade 90.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

In music generation, what might an RNN be trained to predict?

Next note or chord


None of the given options
45804
Next instrument
Next song genre
Next album cover

The correct answer is: Next note or chord

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which problem arises when training RNNs on long sequences?

Underfitting
High bias
All of the given options
Overfitting
Vanishing or exploding gradients

45804
The correct answer is: Vanishing or exploding gradients

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005351&cmid=3690 1/4
1/22/24, 12:38 AM Sequence Generation with RNNs - Post Quiz: Attempt review

Question 3
Incorrect

Mark 0.00 out of 1.00

In sequence generation tasks, what is the primary input to an RNN at each time step?

None of the given options


Current weight
Previous output
Previous error
Current input 

The correct answer is: Previous output

Question 4
Correct

Mark 1.00 out of 1.00


45804
Why might one use GRU over LSTM?

GRU is always more accurate


LSTM can't handle sequences
None of the given options
LSTM is outdated
GRU is simpler and sometimes faster

45804
The correct answer is: GRU is simpler and sometimes faster

Question 5
Correct

Mark 1.00 out of 1.00

What is the primary difference between LSTM and GRU?

LSTM is for sequences, GRU is for images


LSTM is faster, GRU is slower

45804
LSTM has 3 gates, GRU has 2
LSTM has input, forget, and output gates; GRU has reset and update gates
LSTM is older, GRU is newer

The correct answer is: LSTM has input, forget, and output gates; GRU has reset and update gates

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005351&cmid=3690 2/4
1/22/24, 12:38 AM Sequence Generation with RNNs - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a typical use case for RNNs?

Time series prediction


None of the given options
Text generation
Speech recognition
Image classification

The correct answer is: Image classification

Question 7
Correct

Mark 1.00 out of 1.00


45804
How do RNNs handle variable-length sequences in NLP?

By skipping them
Through padding and truncation
None of the given options
By changing the network size
They don't

45804
The correct answer is: Through padding and truncation

Question 8
Correct

Mark 1.00 out of 1.00

Which RNN architecture uses a reset and update gate?

GRU
LSTM

45804
None of the given options
Simple RNN
Bidirectional RNN

The correct answer is: GRU

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005351&cmid=3690 3/4
1/22/24, 12:38 AM Sequence Generation with RNNs - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

What is the main advantage of LSTM over basic RNN?

Handling long-term dependencies


More layers
Lower computational cost
Faster computation
None of the given options

The correct answer is: Handling long-term dependencies

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which of the following is a common application of RNNs in NLP?

Image generation
Face recognition
Object detection
Image classification
Text generation

The correct answer is: Text generation


45804
◄ Sequence Generation with RNNs

Jump to...

Case Study - Sentiment Analysis with RNNs ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005351&cmid=3690 4/4
1/21/24, 2:26 AM Sequence Generation with RNNs - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Sequence Generation with RNNs

Quiz review
Started on Sunday, 21 January 2024, 2:19 AM
State Finished
Completed on Sunday, 21 January 2024, 2:26 AM
Time taken 7 mins 41 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What is the key advantage of using LSTMs over basic RNNs in sequence generation tasks?

Less prone to overfitting


Simpler architecture
45804
Lower computational cost
Ability to remember long-term dependencies
Faster training speeds

The correct answer is: Ability to remember long-term dependencies

Question 2
Correct

Mark 1.00 out of 1.00 45804


In NLP, what does RNNs help to predict?

Next song note


None of the options given
Next video frame
Next image
Next word

The correct answer is: Next word 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990188&cmid=3689 1/4
1/21/24, 2:26 AM Sequence Generation with RNNs - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

During the training of RNNs for sequence generation, what is the common technique used to mitigate the vanishing gradient problem?

Dropout
Batch normalization
Gradient clipping
L1 regularization
Data augmentation

The correct answer is: Gradient clipping

Question 4
Correct

Mark 1.00 out of 1.00


45804
Which RNN architecture utilizes update and reset gates to manage memory?

GRU
LSTM
Echo State Network
Hopfield Network
Bidirectional RNN

The correct answer is: GRU


45804
Question 5
Correct

Mark 1.00 out of 1.00

In the context of natural language processing, how are RNNs typically utilized for machine translation?

For clustering text data


As a replacement for CNNs

45804
As discriminators in GANs
For image classification
Encoding the input sequence and decoding the output sequence

The correct answer is: Encoding the input sequence and decoding the output sequence

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990188&cmid=3689 2/4
1/21/24, 2:26 AM Sequence Generation with RNNs - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

When using RNNs for music generation, what does each neuron in the output layer typically represent?

A specific instrument
A time step in the generated sequence
A frequency band
A possible note or rest in the musical vocabulary
A note in the C major scale

The correct answer is: A possible note or rest in the musical vocabulary

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which of the following is NOT a type of RNN architecture?

GRU
Simple RNN
LSTM
CNN
Bidirectional RNN

The correct answer is: CNN


45804
Question 8
Correct

Mark 1.00 out of 1.00

Which problem in RNNs does LSTM help to address?

Bias
Overfitting

45804
Vanishing gradient
High variance
All of the options given

The correct answer is: Vanishing gradient

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990188&cmid=3689 3/4
1/21/24, 2:26 AM Sequence Generation with RNNs - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

RNNs are primarily used for which type of data?

Sequential
Audio
Tabular
Image
None of the options given

The correct answer is: Sequential

Question 10
Correct

Mark 1.00 out of 1.00


45804
What does RNN stand for?

Recurrent Neural Network


None of the options given
Recursive Neural Network
Regular Neural Network
Random Neural Network

45804
The correct answer is: Recurrent Neural Network

◄ CASE STUDY - GANS - CIFAR - Quiz

Jump to...

Sequence Generation with RNNs ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990188&cmid=3689 4/4
1/22/24, 1:12 AM Transformers and Attention Mechanisms - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Transformers and Attention Mechanisms

Quiz review
Started on Monday, 22 January 2024, 1:07 AM
State Finished
Completed on Monday, 22 January 2024, 1:12 AM
Time taken 4 mins 59 secs
Marks 9.00/10.00
Grade 90.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What is the main difference between pre-training and fine-tuning in Transformers?

45804
Pre-training uses smaller models
None of the options given
Fine-tuning is done without labeled data
Both are done simultaneously
Pre-training is on a large corpus and fine-tuning is task-specific

The correct answer is: Pre-training is on a large corpus and fine-tuning is task-specific

Question 2
Correct

Mark 1.00 out of 1.00 45804


What is the primary task BERT is designed for?

Text generation
Language translation
Image generation
None of the options given
Bidirectional understanding of text

45804
The correct answer is: Bidirectional understanding of text

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005705&cmid=3692 1/4
1/22/24, 1:12 AM Transformers and Attention Mechanisms - Post Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which Transformer model is known for generating coherent paragraphs of text?

BERT
GPT
DALL·E
T5
Image GPT

The correct answer is: GPT

Question 4
Incorrect

Mark 0.00 out of 1.00


45804
Which model can be used for both image and text tasks?

BERT
None of the options given
DALL·E 
T5
GPT

45804
The correct answer is: None of the options given

Question 5
Correct

Mark 1.00 out of 1.00

Which of the following models is designed for image generation?

DALL·E
GPT

45804
None of the options given
T5
BERT

The correct answer is: DALL·E

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005705&cmid=3692 2/4
1/22/24, 1:12 AM Transformers and Attention Mechanisms - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

In the context of Transformers, what does "seq to seq" stand for?

Sequence to Sequence
None of the options given
Sequence training
Sequential training
Sequential to Sequential

The correct answer is: Sequence to Sequence

Question 7
Correct

Mark 1.00 out of 1.00


45804
In sequence-to-sequence tasks, why is attention important?

All of the options given


It helps the model focus on relevant parts of the input
It speeds up computation
It reduces overfitting
It simplifies the model

45804
The correct answer is: It helps the model focus on relevant parts of the input

Question 8
Correct

Mark 1.00 out of 1.00

For which task might you use a Transformer to generate a concise summary of a long article?

Question Answering
Summarization

45804
Translation
None of the options given
Image Classification

The correct answer is: Summarization

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005705&cmid=3692 3/4
1/22/24, 1:12 AM Transformers and Attention Mechanisms - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which mechanism allows Transformers to weigh the importance of different words in a sequence?

Self Attention Mechanism


CNN layers
LSTM cells
None of the options given
RNN cells

The correct answer is: Self Attention Mechanism

Question 10
Correct

Mark 1.00 out of 1.00


45804
How does Multi-head attention differ from standard attention?

It is faster
It is only used in GPT
None of the options given
It allows the model to focus on multiple parts of the input simultaneously
It uses fewer parameters

45804
The correct answer is: It allows the model to focus on multiple parts of the input simultaneously

◄ Transformers and Attention Mechanisms

Jump to...

Case Study - Transformers in Machine Translation ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005705&cmid=3692 4/4
1/22/24, 12:50 AM Transformers and Attention Mechanisms - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Transformers and Attention Mechanisms

Quiz review
Started on Monday, 22 January 2024, 12:44 AM
State Finished
Completed on Monday, 22 January 2024, 12:50 AM
Time taken 5 mins 16 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

What is the first step in training a Transformer model for a specific task?

None of the options given


Fine-tuning
45804
Initialization
Backpropagation
Pre-training

The correct answer is: Pre-training

Question 2
Correct

Mark 1.00 out of 1.00 45804


Which application showcases the use of Transformers in image tasks?

Named entity recognition


Text summarization
Sequence alignment
Image generation using DALL·E
Speech recognition

The correct answer is: Image generation using DALL·E 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005502&cmid=3691 1/4
1/22/24, 12:50 AM Transformers and Attention Mechanisms - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

What is the primary component of the Transformer architecture that helps it handle sequences?

CNN
Attention Mechanism
RNN
None of the options given
LSTM

The correct answer is: Attention Mechanism

Question 4
Correct

Mark 1.00 out of 1.00


45804
Why is attention particularly crucial in sequence-to-sequence tasks like translation?

It reduces the model's size


It makes the model more interpretable
It allows the model to focus on relevant parts of the input when producing an output
It speeds up the training process
It ensures the output is of a fixed size

45804
The correct answer is: It allows the model to focus on relevant parts of the input when producing an output

Question 5
Correct

Mark 1.00 out of 1.00

What does the Multi-head attention mechanism in Transformers help with?

None of the options given


Speeding up training

45804
Capturing different types of information from the input
Reducing model size
Improving regularization

The correct answer is: Capturing different types of information from the input

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005502&cmid=3691 2/4
1/22/24, 12:50 AM Transformers and Attention Mechanisms - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

The Transformer architecture introduced the concept of self-attention to handle which primary challenge in sequence modeling?

Improving model robustness


Reducing model size
Capturing dependencies regardless of their distance in the input
Speeding up training
Handling larger input sizes

The correct answer is: Capturing dependencies regardless of their distance in the input

Question 7
Correct

Mark 1.00 out of 1.00


45804
In the context of Transformers for language translation, what does the encoder primarily focus on?

Reducing the sequence length


Handling attention mechanisms
Generating the final translation
Decoding the target language
Processing and representing the source language

45804
The correct answer is: Processing and representing the source language

Question 8
Correct

Mark 1.00 out of 1.00

What is the primary advantage of pretraining a Transformer on a large corpus before fine-tuning on a specific task?

It allows the model to leverage general language understanding


It makes the model more robust to adversarial attacks

45804
It reduces the risk of overfitting
It speeds up the fine-tuning process
It makes the model smaller

The correct answer is: It allows the model to leverage general language understanding

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005502&cmid=3691 3/4
1/22/24, 12:50 AM Transformers and Attention Mechanisms - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which Transformer model is specifically designed for language translation?

Image GPT
T5
GPT
BERT
DALL·E

The correct answer is: T5

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which of the following is NOT a sequence-to-sequence task?

Summarization
Image Classification
Translation
Question Answering
None of the options given

The correct answer is: Image Classification


45804
◄ Sentiment Analysis with RNNs - Case study

Jump to...

Transformers and Attention Mechanisms ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=1005502&cmid=3691 4/4
1/21/24, 1:34 AM Variational Autoencoders - Post Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Variational Autoencoders

Quiz review
Started on Sunday, 21 January 2024, 1:28 AM
State Finished
Completed on Sunday, 21 January 2024, 1:33 AM
Time taken 5 mins 3 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Which optimization technique is commonly used with VAEs?

None of the given options 45804


Stochastic gradient descent (SGD)
Principal component analysis
Simulated annealing
Genetic algorithms

The correct answer is: Stochastic gradient descent (SGD)

Question 2
Correct

Mark 1.00 out of 1.00 45804


In which application might VAEs be used to enhance image quality?

Social media photo filters


Text generation
None of the given options
Medical imaging
Video streaming

The correct answer is: Medical imaging 45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990015&cmid=3681 1/4
1/21/24, 1:34 AM Variational Autoencoders - Post Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

How do VAEs differ from traditional autoencoders?

VAEs are more accurate


VAEs are faster
VAEs introduce randomness via a probabilistic layer
VAEs are simpler
VAEs use supervised learning

The correct answer is: VAEs introduce randomness via a probabilistic layer

Question 4
Correct

Mark 1.00 out of 1.00


45804
Why is the reparameterization trick important in VAEs?

None of the given options


It simplifies the model architecture
It increases model efficiency
It allows backpropagation through random nodes
It reduces overfitting

45804
The correct answer is: It allows backpropagation through random nodes

Question 5
Correct

Mark 1.00 out of 1.00

Why is variational inference used in VAEs?

To improve model accuracy


To speed up training

45804
To reduce model size
To approximate intractable posterior distributions
None of the given options

The correct answer is: To approximate intractable posterior distributions

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990015&cmid=3681 2/4
1/21/24, 1:34 AM Variational Autoencoders - Post Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

What do VAEs use to generate a distribution over latent variables?

Variational inference
Backpropagation
Gradient descent
None of the given options
Transfer learning

The correct answer is: Variational inference

Question 7
Correct

Mark 1.00 out of 1.00


45804
Autoencoders primarily focus on which aspect of data?

Classification
Generation
Reconstruction
Filtering
Clustering

The correct answer is: Reconstruction


45804
Question 8
Correct

Mark 1.00 out of 1.00

In which application can VAEs detect unusual patterns?

Fashion design
Anomaly detection in industrial equipment

45804
Music composition
Text generation
Face generation for video games

The correct answer is: Anomaly detection in industrial equipment

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990015&cmid=3681 3/4
1/21/24, 1:34 AM Variational Autoencoders - Post Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a typical use case for VAEs?

Medical imaging enhancement


Face generation for video games
Real-time speech translation
Anomaly detection in industrial equipment
Fashion design

The correct answer is: Real-time speech translation

Question 10
Correct

Mark 1.00 out of 1.00


45804
Which of the following is a key component of the VAE loss function?

KL divergence
Accuracy
Cross-entropy only
Precision
Mean squared error only

The correct answer is: KL divergence


45804
◄ Variational Autoencoders

Jump to...

Case Study - Variational Auto Encoder ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=990015&cmid=3681 4/4
1/21/24, 1:16 AM Variational Autoencoders - Pre Quiz: Attempt review


 Dashboard / Primer 2.0 - App Dev / Stage 1 / Gen AI / Variational Autoencoders

Quiz review
Started on Sunday, 21 January 2024, 1:09 AM
State Finished
Completed on Sunday, 21 January 2024, 1:16 AM
Time taken 7 mins 26 secs
Marks 10.00/10.00
Grade 100.00 out of 100.00

Question 1
Correct

Mark 1.00 out of 1.00

Why is the reparameterization trick crucial in training VAEs?

45804
It allows backpropagation through stochastic nodes
It reduces the need for labeled data
It reduces the model's complexity
It increases the model's accuracy
It speeds up the training process

The correct answer is: It allows backpropagation through stochastic nodes

Question 2
Correct

Mark 1.00 out of 1.00 45804


Reparameterization trick is used to...

Deal with the non-differentiability of sampling in VAEs


None of the given options
Speed up training
Reduce model size
Improve model accuracy

45804
The correct answer is: Deal with the non-differentiability of sampling in VAEs

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989934&cmid=3680 1/4
1/21/24, 1:16 AM Variational Autoencoders - Pre Quiz: Attempt review

Question 3
Correct

Mark 1.00 out of 1.00

Which application does NOT typically use VAEs?

Medical imaging enhancement


Anomaly detection in industrial equipment
Fashion design
Face generation for video games
Text summarization

The correct answer is: Text summarization

Question 4
Correct

Mark 1.00 out of 1.00


45804
In which application might you use a VAE for generating new, coherent samples?

Text translation
Designing virtual fashion items
Time series forecasting
Speech recognition
Image classification

45804
The correct answer is: Designing virtual fashion items

Question 5
Correct

Mark 1.00 out of 1.00

Which of the following is NOT a type of autoencoder?

Variational autoencoder
Supervised autoencoder

45804
Denoising autoencoder
Sparse autoencoder
Contractive autoencoder

The correct answer is: Supervised autoencoder

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989934&cmid=3680 2/4
1/21/24, 1:16 AM Variational Autoencoders - Pre Quiz: Attempt review

Question 6
Correct

Mark 1.00 out of 1.00

What does VAE stand for?

Vectorized Autoencoder
Virtual Autoencoder
Variable Autoencoder
None of the given options
Variational Autoencoder

The correct answer is: Variational Autoencoder

Question 7
Correct

Mark 1.00 out of 1.00


45804
Which component of the VAE loss function ensures the latent variables adhere to a standard distribution?

Mean squared error


Cross-entropy
KL divergence
Absolute error
Hinge loss

The correct answer is: KL divergence


45804
Question 8
Correct

Mark 1.00 out of 1.00

What is the primary role of autoencoders in generative modeling?

Data classification
Clustering

45804
Image recognition
Regression
Data compression and reconstruction

The correct answer is: Data compression and reconstruction

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989934&cmid=3680 3/4
1/21/24, 1:16 AM Variational Autoencoders - Pre Quiz: Attempt review

Question 9
Correct

Mark 1.00 out of 1.00

Why are autoencoders considered generative models?

They are used for supervised learning


They can reconstruct and generate data similar to the input
They always reduce data dimensionality
They are only used for image data
They are a type of neural network

The correct answer is: They can reconstruct and generate data similar to the input

Question 10
Correct

Mark 1.00 out of 1.00


45804
In the context of Variational Autoencoders (VAEs), what does variational inference help achieve?

Faster training speeds


Reduction of model parameters
Direct computation of posterior distributions
Improved image resolution
Approximation of complex posterior distributions

45804
The correct answer is: Approximation of complex posterior distributions

◄ Introduction to Generative Models - Post Quiz

Jump to...

Variational Autoencoders ►

45804

https://accenturelearning.tekstac.com/mod/quiz/review.php?attempt=989934&cmid=3680 4/4

You might also like