Domain 2: Gen AI Fundamentals 24%

Question 1

____ is a subset of deep learning. Like deep learning, this is a multipurpose technology that helps to generate new original content rather than finding or classifying existing content.

Answer 1

Generative AI

Question 2

_____ looks for statistical patterns in modalities, such as natural language and images.

Answer 2

Gen AI foundational models

Question 3

_____ are very large and complex neural network models with billions of parameters that are learned during the training phase or pre-training.

Answer 3

Gen AI foundational models

Question 4

The more parameters a model has, the more _____ it has, so the model can perform more advanced tasks.

Answer 4

memory

Question 5

Gen AI models are built with _____, _____, _____,and_____ all working together.

Answer 5

neural networks, system resources, data, and prompts

Question 6

The current core element of generative AI is the _____.

Answer 6

transformer network

Question 7

_____ are pre-trained on massive amounts of the text data from the internet, and they can use this pre-training process to build up a broad knowledge base.

Answer 7

Large Language Models (LLMs)

Question 8

A ____is a natural language text that requests the generative AI to perform a specific task.

Answer 8

prompt

Question 9

The process of reducing the size of one model (known as the teacher) into a smaller model (known as the student) that emulates the original model’s predictions as faithfully as possible.

Answer 9

distillation

Question 10

A prompt that contains more than one example demonstrating how the large language model should respond.

Answer 10

few-shot prompting

Question 11

A second, task-specific training pass performed on a pre-trained model to refine its parameters for a specific use case.

Answer 11

fine tuning

Question 12

A form of fine-tuning that improves a generative AI model’s ability to follow instructions, which involves training a model on a series of instruction prompts, typically covering a wide variety of tasks. The resulting model generates useful responses to zero-shot prompts across a variety of tasks.

Answer 12

instruction tuning

Question 13

An algorithm for performing parameter efficient tuning that fine-tunes only a subset of a large language model’s parameters.

Answer 13

Low-Rank Adaptability (LoRA)

Question 14

A system that picks the ideal model for a specific inference query.

Answer 14

model cascading

Question 15

The algorithm that determines the ideal model for inference in model cascading. It is typically a machine learning model that gradually learns how to pick the best model for a given input, and it could sometimes be a simpler, non-machine learning algorithm.

Answer 15

model router

Question 16

A prompt that contains one example demonstrating how the large language model should respond. For example, the following prompt contains one example showing a large language model how it should answer a query.

Answer 16

one-shot prompting

Question 17

A set of techniques to fine-tune a large pre-trained language model (PLM) more efficiently than full fine-tuning. It typically fine-tunes far fewer parameters than full fine-tuning, yet generally produces a large language model that performs as well (or almost as well) as a large language model built from full fine-tuning.

Answer 17

parameter-efficient tuning

Question 18

Models or model components (such as an embedding vector) that have already been trained. Sometimes, you’ll feed pre-trained embedding vectors into a neural network. Other times, your model will train the embedding vectors themselves rather than rely on the pre-trained embeddings.

Answer 18

pre-trained model

Question 19

The initial training of a model on a large dataset. Some models are clumsy giants and must typically be refined through additional training.

Answer 19

pre-training

Question 20

Any text entered as input to a large language model to condition the model to behave in a certain way. These can be as short as a phrase or arbitrarily long (for example, the entire text of a novel).

Answer 20

prompt

Question 21

A capability of certain models that enables them to adapt their behavior in response to arbitrary text input (prompts). In the paradigm, a large language model responds to a prompt by generating text.

Answer 21

prompt-based learning

Question 22

The art of creating prompts that elicit the desired responses from a large language model. Humans perform. Writing well-structured prompts is an essential part of ensuring useful responses from a large language model.

Answer 22

prompt engineering

Question 23

Using feedback from human raters to improve the quality of a model’s responses. The system can then adjust its future responses based on that feedback.

Answer 23

Reinforcement Learning from Human Feedback (RLHF)

Question 24

An optional part of a prompt that identifies a target audience for a generative AI model’s response. Without it, a large language model provides an answer that may or may not be useful for the person asking the questions. With it, a large language model can answer in a way that’s more appropriate and more helpful for a specific target audience.

Answer 24

role prompting

Question 25

A technique for tuning a large language model for a particular task, without resource intensive fine-tuning. Instead of retraining all the weights in the model, this automatically adjusts a prompt to achieve the same goal.

Answer 25

soft prompt tuning

Question 26

A hyperparameter that controls the degree of randomness of a model's output. Higher result in more random output, while lower result in less random output.

Answer 26

temperature

Question 27

A prompt that does not provide an example of how you want the large language model to respond.

Answer 27

zero-shot prompting

Question 28

The process that a trained machine learning model uses to draw conclusions from brand-new data.

Answer 28

inference

Question 29

When the purpose of the prompt is finished.

Answer 29

completion

Question 30

The amount of textual information that the AI can take into account at any given time when processing language

Answer 30

context window

Question 31

The smallest units of text that an AI model processes.

Answer 31

tokens

Question 32

This breaks down text into these units to make it manageable for computational models.

Answer 32

tokenizer

Question 33

Providing examples inside the context window is called _____. With this, you can help LLMs learn more about the task being asked by including examples or additional data in the prompt.

Answer 33

in-context learning

Question 34

The input that you sent into your generative model is called the _____, which consists of instructions and content.

Answer 34

prompt

Question 35

T/F: The larger a model is, the more likely it is to work without additional in-context learning or further training. Because the model's capability increases with size, it has supported the development of larger and larger models.

Answer 35

True

Question 36

LLMs encode a deep statistical representation of a language. This understanding is developed during the _____ phase when the model learns from vast amounts of unstructured data.

Answer 36

pre-training

Question 37

During _____, the model weights get updated to minimize the loss of the training objective, and the encoder generates an embedding or vector representation for each token.

Answer 37

pre-training

Question 38

Why do you need to process training data collected from public sites?

Answer 38

to increase quality, address bias, or remove other harmful content.

Question 39

_____ work with one data modality. LLMs are an example of this type of generative AI because the input and the output, or completion, are text.

Answer 39

Unimodal models

Question 40

_____ is adding another modality such as image, video, or audio. These models can understand diverse data sources and can provide more robust forecasts.

Answer 40

Multimodal

Question 41

These models can process and generate multiple types of data, but they can also do this type of operation in combination with each other. This collaborative capability adds cross-model reasoning, translation, search, and creation that more closely mirrors human intelligence.

Answer 41

Multimodal models

Question 42

_____ use cases are marketing, image captioning, product design, customer service, chatbots, and avatars.

Answer 42

Multimodal generative AI

Question 43

1. image captioning, where the model is generating text descriptions of images 2. visual question answering, where the model answers questions about image content 3. text to image synthesis, which is generating images from textual descriptions.

Answer 43

examples of multimodal tasks

Question 44

T/F: Models such as DALL-E, Stable Diffusion, and Midjourney can create realistic and diverse images from natural language prompts. There are also diffusion models, which support a variety of tasks for multimodal models such as image generation, upscaling, and inpainting.

Answer 44

True

Question 45

_____ are a class of generative models that learn to reverse a gradual noising process. _____-based architectures offer a higher degree of control in quality and diversity of images generated.

Answer 45

Diffusion models

Question 46

The _____ process begins by sampling from a basic, usually Gaussian, distribution. This initial simple sample undergoes a series of reversible, incremental modifications, where each step introduces a controlled amount of complexity through a Markov chain

Answer 46

forward diffusion

Question 47

The process of training a neural network to recover the original data by reversing the noising process. Denoising an image to reconstruct the original image.

Answer 47

reverse diffusion

Question 48

_____ is a deep learning model used for converting text to images. It can generate high-quality, photo-realistic images. Instead of using the pixel space of the image, it uses a reduced definition latent space.

Answer 48

Stable Diffusion

Question 49

T/F: Diffusion models tend to produce higher quality outputs with more diversity and consistency, and they're more stable and easier to train.

Answer 49

True

Question 50

Stable Diffusion for image generation, Whisper for speech recognition and translation, and AudioLM for audio generation.

Answer 50

Examples of diffusion models

Question 51

_____ are a type of generative AI that can be applied to many different problem domains or tasks without being fine-tuned.

Answer 51

Large language models, LLMs

Question 52

1. Generates text 2. Text summarization 3. Info extraction 4. Question answering 5. Classification 6. Identifying harmful content 7. Chatbots 8. Dev tool for source code generation

Answer 52

Use cases for Gen AI

Question 53

_____ and _____ offer pre-trained models for text, image, audio generation, that could be fine-tuned for specific use cases.

Answer 53

Amazon Bedrock and Amazon Titan

Question 54

_____ and _____ formerly known as Amazon CodeWhisper, support code generation and completion.

Answer 54

SageMaker and Amazon Q Developer

Question 55

_____ and _____ offer virtual production and 3D content creation.

Answer 55

Amazon Nimble Studio and Amazon Sumerian

Question 56

_____ is an application of generative AI to accelerate software development. Models can generate functional code snippets, and even entire programs from natural language descriptions or examples, to help automate routine programming tasks, suggest code completions, and even translate code between different languages.

Answer 56

Code generation

Question 57

_____ generates real-time code suggestions, that range from snippets to full functions, based on your comments and existing code. AWS handles the underlying infrastructure, data management, model training, and inference. You can focus on your specific use cases and applications.

Answer 57

Amazon Q Developer

Question 58

A _____ is a deep learning architecture. It trains two neural networks to compete against each other to generate more authentic new data from a given training dataset. Called adversarial because it trains two different networks and pits them against each other. One network generates new data by taking an input data sample and modifying it as much as possible. The other network tries to predict whether the generated data output belongs in the original dataset.

Answer 58

generative adversarial network (GAN)

Question 59

A _____ is a type of neural network that learns to reproduce its input, and also map data to latent space.

Answer 59

variational autoencoder (VAE)

Question 60

_____ are a type of neural network architecture that transforms or changes an input sequence into an output sequence. They do this by learning context and tracking relationships between sequence components.

Answer 60

Transformers

Question 61

The _____ stage of the AI cycle is to define objectives, collect data, process data, select your model, and train and develop your model.

Answer 61

first

Question 62

The _____ stage of the AI cycle is to develop your model by using feature engineering, building, testing, validating, optimizing, and scaling.

Answer 62

second

Question 63

The _____ stage of the AI cycle to deploy and maintain, which includes model evaluation, deployment, feedback, updates, security, and scalability.

Answer 63

third

Question 64

Foundation model lifecycle

Answer 64

1. data selection 2. model selection 3. pre-training 4. fine-tuning 5. evaluation 6. deployment 7. feedback

Question 65

1. scope 2. train model or work with existing 3. assess model's performance and additional training 4. reinforcement learning from human feedback 5. evaluation (metrics) 6. prompt engineering then fine tune 7. revisit and evaluate 8. deploy 9. consider needed infrastructure

Answer 65

Foundation model lifecycle

Question 66

What are the advantages of Gen AI?

Answer 66

adaptability, responsiveness, and simplicity

Question 67

_____ are incorrect or misleading results that AI models generate. These errors can be caused by a variety of factors, including insufficient training data, incorrect assumptions made by the model, or biases in the data used to train the model.

Answer 67

AI hallucinations

Question 68

T/F: The higher the interpretability of a machine learning model, the easier it is to comprehend the model's predictions.

Answer 68

True

Question 69

_____ can be applied to interpret models that have low complexity or simple relationships between the input variables and the predictions. The simple relationship between the inputs, input variables, output results, and predictions in high model interpretability can lead to lower performance.

Answer 69

Intrinsic analysis

Question 70

Cross-domain performance, efficiency, conversion rate, average revenue per user, accuracy, customer lifetime value

Answer 70

business metrics for Gen AI

Question 71

_____ are trained on huge, unlabeled, broad datasets and they underpin the capabilities of generative AI. As a result, they are considerably larger than traditional machine learning models, which are generally used for more specific functions. They are used as the baseline starting point for developing and creating models, and they can be used to interpret and understand language, have conversational messaging, and create and generate images.

Answer 71

Foundation models

Question 72

AWS provides Amazon's business metric analysis ML solution which uses _____ and _____.

Answer 72

Amazon Lookout for Metrics and Amazon Forecast

Question 73

FM output quality metrics are:

Answer 73

relevance, accuracy, coherence, and appropriateness

Question 74

Task completion rates and reduction in manual efforts making a direct contribution to operational productivity

Answer 74

Efficiency metrics

Question 75

Accuracy and credibility metrics:

Answer 75

low error rate

Question 76

Accessibility, lower barrier to entry, efficiency, cost-effectiveness, speed to market, and ability to meet business objectives.

Answer 76

advantages of using AWS generative AI services to build applications

Question 77

The process of fine-tuning and training a pre-trained model on a new dataset without training from scratch is also known as _____.

Answer 77

transfer learning

Question 78

T/F: You can accelerate the learning for some types of models by using transfer learning, then use a pre-trained model as your starting point for the training of a new dataset.

Answer 78

True

Question 79

_____ helps to find projects that are already built and are quick builds with datasets, models, algorithm types, and solutions based on industry best practices.

Answer 79

SageMaker JumpStart

Question 80

_____ has specialized hardware and associated firmware that are designed to enforce security restrictions. These restrictions are to ensure that no one can access your workloads or data running on Amazon Elastic Compute Cloud, Amazon EC2, instances. This protection applies to all Nitro-based applications and instances.

Answer 80

The AWS Nitro System

Question 81

At AWS, securing AI infrastructure means _____, such as AI model weights and data processed with those models by any unauthorized person.

Answer 81

zero access to sensitive AI data

Question 82

3 layers of Gen AI stack

Answer 82

bottom: tools for building/training LLMs and FMs middle: access to all the models + tools to build/scale apps top: apps that use LLMs and FMs to write/debug code, generate content, derive insights, and take actions.

Question 83

What are the three critical components of any AI system?

Answer 83

It's input, model, and output.

Question 84

Pricing models for LLMs:

Answer 84

1. host LLMs on your own infrastructure 2. pay by token

Question 85

_____ are the units that vendors use to price calls in their APIs.

Answer 85

Tokens

Question 86

What's the benefit of using pay-by-token?

Answer 86

scalability

Question 87

Why does AWS Global Infrastructure add high availability and fault tolerance?

Answer 87

It's comprised of multiple physically separated and isolated Availability Zones (AZs) within each Region, AWS ensures redundancy and fault tolerance, safeguarding against service disruptions.

Question 88

_____ is a model hub and it helps quickly deploy foundation models that are available within the service and integrate them into your applications. It provides fine-tuning and deploying models, and also helps you get quickly into production to operate at scale.

Answer 88

SageMaker JumpStart

Question 89

_____ is a managed AWS service that lets you use and access a number of different foundation models, FMs, using APIs. It adds the capability to import custom weights for supportive model architectures, and serve the custom model by using on-demand mode. Lets you experiment by running model inference against different base foundation models that are supported within the service to help you align your use cases with the highest accuracy.

Answer 89

Amazon Bedrock

Question 90

The _____ foundation model provided by Amazon acts as a general-purpose foundation model and is a great option if you require tax generation capabilities.

Answer 90

Amazon Titan

Question 91

_____ is a playground built on Amazon Bedrock. You can use it to build generative AI applications to learn fundamental techniques, such as understanding how a foundation model responds to different prompts. Some examples of applications that you can build with this are creating playlists, trivia games, recipes, and more.

Answer 91

PartyRock

Question 92

Remember that with generative AI, you can use _____, and that data is stored as embeddings. These embeddings are vectors that can be compressed, stored and indexed for advanced searches.

Answer 92

vector databases