1.What is Generative AI Studio?
 
(1) A machine learning model that is trained on text only.
 
(2) A technology that lets you code programming languages without learning them.
 
(3) A type of artificial intelligence that writes emails for you.
 
(4) A tool that helps you use Generative AI capabilities in your application. 정답
 
2.How does generative AI generate new content?
 
(1) It is programmed based on predetermined algorithms that can not be altered.
 
(2) It learns from a massive amount of existing content. 정답
 
(3) It is a random process.
 
(4) The training leads to a foundation model that cannot be further tuned with a new dataset.
 
3.What is a prompt?
 
(1) A prompt is a long piece of text that explains how a large language model generates text.
 
(2) A prompt is a piece of text that is used to evaluate a large language model.
 
(3) A prompt is a short piece of text that is used to guide a large language model to generate content. 정답
 
(4) A prompt is a piece of text that is used to train a large language model.
 
4. Which of the following is a type of prompt that allows a large language model to perform a task with only a few examples? 틀림;
 
(1) Few-shot prompt
 
(2) One-shot prompt
 
(3) Zero-shot prompt
 
(4) Unsupervised prompt
 
5. Which of the following is the best way to generate more creative or unexpected content by adjusting the model parameters in Generative AI Studio?
 
(1) Setting the top K to 1
 
(2) Setting the temperature to a high value 정답
 
(3) Setting the top P to 25%
 
(4) Setting the temperature to a low value

 

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Create Image Captioning Models: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Transformer Models and BERT Model: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Attention Mechanism: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11

1. What is the purpose of the decoder in an encoder-decoder model?

(1) To store the output data

(2) To learn the relationship between the input and output data

(3) To extract information from the input data

(4) To generate output data from the information extracted by the encoder 정답


2.What is the purpose of the encoder in an encoder-decoder model?

(1) To generate text captions for the image.

(2) To extract information from the image. 정답

(3) To translate text from one language to another.

(4) To answer your questions in an informative way, even if they are open ended, challenging, or strange.


3. What is the name of the model that is used to generate text captions for images?

(1) Bidirectional Encoder Representations from Transformers (BERT) model

(2) Encoder-decoder model 정답

(3) Image generation model

(4) Image classification model


4. What is the goal of the image captioning task?

(1) To answer your questions in an informative way, even if they are open ended, challenging, or strange.

(2) To write different creative text formats, like poems, code, scripts, musical pieces, email, letters, etc.

(3) To translate text from one language to another

(4) To generate text captions for images 정답

 

5.What is the name of the dataset the video uses to train the encoder-decoder model?

(1) ImageNet dataset

(2) COCO dataset 정답

(3) Fashion-MNIST dataset

(4) MNIST dataset


6.What is the purpose of the attention mechanism in an encoder-decoder model?

(1) To translate text from one language to another.

(2) To extract information from the image.

(3) To allow the decoder to focus on specific parts of the image when generating text captions. 정답

(4) To generate text captions for the image.

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Generative AI Studio: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Transformer Models and BERT Model: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Attention Mechanism: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
1. What are the two sublayers of each encoder in a Transformer model?
 
Convolution and pooling
 
Embedding and classification
 
Recurrent and feedforward
 
Self-attention and feedforward 정답 
2. What is the name of the language modeling technique that is used in Bidirectional Encoder Representations from Transformers (BERT)?
 
Recurrent Neural Network (RNN)
 
Long Short-Term Memory (LSTM)
 
Transformer 정
 
Gated Recurrent Unit (GRU)
3. What does fine-tuning a BERT model mean?
 
Training the hyper-parameters of the models on a specific task
 
Training the model on a specific task by using a large amount of unlabeled data
 
Training the model on a specific task and not updating the pre-trained weights
 
Training the model and updating the pre-trained weights on a specific task by using labeled data 정답 
4. What kind of transformer model is BERT?
 
Encoder-decoder model
 
Recurrent Neural Network (RNN) encoder-decoder model
 
Decoder-only model
 
Encoder-only model 정답 
5. What are the encoder and decoder components of a transformer model?
 
The encoder ingests an input sequence and produces a sequence of hidden states. The decoder takes in the hidden states from the encoder and produces an output sequence. 정답
 
The encoder ingests an input sequence and produces a sequence of tokens. The decoder takes in the tokens from the encoder and produces an output sequence.
 
The encoder ingests an input sequence and produces a single hidden state. The decoder takes in the hidden state from the encoder and produces an output sequence.
 
The encoder ingests an input sequence and produces a sequence of images. The decoder takes in the images from the encoder and produces an output sequence.
6. What is the attention mechanism?
 
A way of identifying the topic of a sentence
 
A way of determining the similarity between two sentences
 
A way of predicting the next word in a sentence 정답
 
A way of determining the importance of each word in a sentence for the translation of another sentence
7. What is a transformer model?
 
A deep learning model that uses self-attention to learn relationships between different parts of a sequence. 정답
 
A natural language processing model that uses convolutions to learn relationships between different parts of a sequence.
 
A machine learning model that uses recurrent neural networks to learn relationships between different parts of a sequence.
 
A computer vision model that uses fully connected layers to learn relationships between different parts of an image.
8. What are the three different embeddings that are generated from an input sentence in a Transformer model?
 
Recurrent, feedforward, and attention embeddings
 
Embedding, classification, and next sentence embeddings
 
Convolution, pooling, and recurrent embeddings
 
Token, segment, and position embeddings 정답
9. BERT is a transformer model that was developed by Google in 2018. What is BERT used for?
 
It is used to solve many natural language processing tasks, such as question answering, text classification, and natural language inference. 정답
 
It is used to diagnose and treat diseases.
 
It is used to train other machine learning models, such as Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks.
 
It is used to generate text, translate languages, and write different kinds of creative content.

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Generative AI Studio: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Create Image Captioning Models: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Attention Mechanism: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
1.What is the name of the machine learning architecture that can be used to translate text from one language to another?
 
Convolutional neural network (CNN)
 
Long Short-Term Memory (LSTM)
 
Neural network
Encoder-decoder 정답

That's correct!

2. What is the advantage of using the attention mechanism over a traditional recurrent neural network (RNN) encoder-decoder?
 
The attention mechanism requires less CPU threads than a traditional RNN encoder-decoder.
 
The attention mechanism is faster than a traditional RNN encoder-decoder.정답
The attention mechanism lets the decoder focus on specific parts of the input sequence, which can improve the accuracy of the translation.
 
The attention mechanism is more cost-effective than a traditional RNN encoder-decoder.

That's correct!

3. How does an attention model differ from a traditional model?
 
The decoder does not use any additional information.
Attention models pass a lot more information to the decoder. 정답
 
The decoder only uses the final hidden state from the encoder.
 
The traditional model uses the input embedding directly in the decoder to get more context.

That's correct!

4. What are the two main steps of the attention mechanism?
 
Calculating the attention weights and generating the output word
 
Calculating the context vector and generating the attention weights
 
Calculating the context vector and generating the output word
Calculating the attention weights and generating the context vector 정답

That's correct!

5. What is the name of the machine learning technique that allows a neural network to focus on specific parts of an input sequence?
 
Convolutional neural network (CNN)
 
Long Short-Term Memory (LSTM)
 
Encoder-decoder
Attention mechanism 정답

That's correct!

6. What is the purpose of the attention weights?
 
To calculate the context vector by averaging words embedding in the context.
To assign weights to different parts of the input sequence, with the most important parts receiving the highest weights. 정답
 
To generate the output word based on the input data alone.
 
To incrementally apply noise to the input data.

That's correct!

7. What is the advantage of using the attention mechanism over a traditional sequence-to-sequence model?
 
The attention mechanism reduces the computation time of prediction.
 
The attention mechanism lets the model learn only short term dependencies.
 
The attention mechanism lets the model formulate parallel outputs.

 

The attention mechanism lets the model focus on specific parts of the input sequence.정답

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Create Image Captioning Models: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Transformer Models and BERT Model: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Generative AI Fundamentals Quiz  (0) 2023.10.11
1.What is the purpose of the decoder in an encoder-decoder architecture?
 
To convert the input sequence into a vector representation
 
To predict the next word in the output sequence
 
To generate the output sequence from the vector representation 정답
 
To learn the relationship between the input and output sequences
2.What are two ways to generate text from a trained encoder-decoder model at serving time?
 
Teacher forcing and beam search
 
Greedy search and beam search 정답
 
Teacher forcing and attention
 
Greedy search and attention
3.What is the purpose of the encoder in an encoder-decoder architecture?
 
To learn the relationship between the input and output sequences
 
To generate the output sequence from the vector representation
 
To predict the next word in the output sequence 정답
 
To convert the input sequence into a vector representation
4.What is the name of the machine learning architecture that takes a sequence of words as input and outputs a sequence of words?
 
Large stream text manipulation
 
Encoder-decoder 정답
 
Regressive neural networking
 
Collaborative natural network
5.What is the difference between greedy search and beam search?
 
Greedy search always selects the word with the highest probability, whereas beam search considers multiple possible words and selects the one with the highest combined probability. 정답
 
Greedy search considers multiple possible words and selects the one with the highest combined probability, whereas beam search always selects the word with the highest probability.
 
Greedy search always selects the word with the lowest probability, whereas beam search considers multiple possible words and selects the one with the lowest combined probability.
 
Greedy search considers multiple possible words and selects the one with the lowest combined probability, whereas beam search always selects the word with the lowest probability.

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Transformer Models and BERT Model: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Attention Mechanism: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Generative AI Fundamentals Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Responsible AI: Quiz  (0) 2023.10.11
1.What is the process of forward diffusion?
 
Start with a clean image and add noise iteratively 정답
 
Start with a noisy image and remove noise randomly
 
Start with a noisy image and remove noise iteratively
 
Start with a clean image and add noise randomly
 
2.What is the goal of diffusion models?
 
To generate images by treating an image as a sequence of vectors
 
To pit two neural networks against each other
 
To encode images to a compressed size, then decode back to the original size
 
To learn the latent structure of a dataset by modeling the way in which data points diffuse through the latent space 정답

 

3. What is the name of the model family that draws inspiration from physics and thermodynamics?
 
Variational autoencoders
 
Diffusion models 정답
Autoregressive models
 
Generative adversarial networks
 
4.Which process involves a model learning to remove noise from images?
 
Sampling
 
Forward diffusion
 
Reverse diffusion 정답
 
GANs
5.What are some challenges of diffusion models?
 
They can be difficult to control.
 
All of the above 정답
 
They can be computationally expensive to train.
 
They can generate images that are not realistic.

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Attention Mechanism: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Generative AI Fundamentals Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Responsible AI: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Large Language Models: Quiz  (0) 2023.10.11
1.Prompt tuning is a technique for:
 
Making large language models more accurate.
 
Improving the output quality of large language models. 정답
 
Making large language models more versatile.
 
Training large language models.
 
2.Which are examples of tasks that Bard code generation can perform? (Select 3)
 
Detect emerging security vulnerabilities
 
Translate code from one language to another 정답
 
Debug your lines of source code 정답
 
Explain your code to you line by line 정답
 
3. What is a prompt?
 
A guide to write code in different programming languages
 
A natural language processing concept that involves text embeddings
 
A text to query or instruct a large language model (LLM) to generate desired output 정답
 
A pre-trained AI model to perform a variety of tasks
4.You are planning a trip to Germany in a few months. You’ve booked a hotel that includes dinner. You want to email the hotel so they are aware of your food preferences. Using gen AI, you can easily generate an email in German using English prompts. Which gen AI training model enables you to do this?
 
Text-to-video
 
Text-to-3D
 
Text-to-image
 
Text-to-text 정답
 
5.According to Google’s AI Principles, bias can enter into the system at only specific points in the ML lifecycle.
 
True 
 
False 정답
 
6. Your company has a large language model (LLM) to help employees with their everyday tasks. You want to teach your colleagues about prompt design. Which of the following is a good example of a well-designed prompt?
 
When does my manager expect me to complete my project?
 
What is the meaning of life?
 
Can you tell me how long a really long paragraph can be before it’s too long?
 
Translate the following sentence into French: Hello, how are you today? 정답
 
7.Google’s approach to responsible AI is based on which of the following commitments? (Select 4) 공부필요 
 
It’s accountable and safe.
 
It eliminates team disagreements.
 
It’s built for everyone.
 
It speeds up product development processes.
 
It’s driven by scientific excellence.
 
It respects privacy.
 
8.Which of the following is a potential use of generative AI?
 
Automatic reading of X-rays
 
A customer service chatbot 정답
 
Predicting future sales
 
A robot that performs mechanical tasks
 
9.What are some benefits of prompt tuning? (Select 2)
 
Generalizes large language models’ (LLMs) commands to conduct versatile tasks
 
Enables large language models (LLMs) to be trained on small amounts of data
 
Enables large language models (LLMs) to adapt to a wide range of tasks 정답
 
Increases large language models’ (LLMs) production of unbiased responses
 
Helps large language models (LLMs) generate more accurate responses 정답
 
10.You want to explain to a colleague how gen AI works. Which of the following would be a good explanation?
 
Gen AI randomly generates new content without any input from existing data.
 
Gen AI learns from existing data and then creates new content that is similar to the data it was trained on. 정답
 
Gen AI uses a set of rules to generate new content that is always unique and original.
 
Gen AI determines the relationship between datasets and classifies data according to existing data sets.
 
11. You want to use machine learning to discover the underlying pattern and group a collection of unlabeled photos into different sets. Which should you use?
 
Supervised learning
 
Unsupervised learning 정답
 
Large language models (LLM)
 
Bard
 
12. Which of the following are examples of pre-training for a large language model (LLM)? (Select 3)
 
Financial forecasting
 
Question answering 정답
 
Document summarization 정답
 
Text classification 정답
 
13.Which of the following is recommended by Google as most important in establishing AI governance?
 
Developing technical tools to evaluate ML models
 
Encouraging people to make decisions based on their own values
 
Creating decision trees to address common ethical situations
 
Building practices around ethical decision-making 정답
 
14.How does generative AI work?
 
It uses a neural network to learn from a large dataset. 정답
 
It uses the internet to repeat answers for common questions.
 
It uses a generic algorithm to evolve a population of models until it finds one that can generate the desired output.
 
It uses a rule-based system to generate output based on a set of predefined rules. 
 
15.You want to organize documents into distinct groups, without predefining the groups. Which type of machine learning model should you use?
 
Discriminative deep learning model 
 
Supervised learning model
 
Natural language processing (NLP) model
 
Unsupervised learning model 정답

 

16.What is machine learning?
 
Algorithms used to describe and create new data
 
Programs or systems that learn from data instead of being explicitly programmed 정답

 

 
Large, general purpose language models
 
Artificial intelligence technology that can produce various types of content 
17. What are the elements of a transformer model at a high-level?
 
Supervised and unsupervised
 
Text and image
 
Encoder and decoder 정답
 
Generative and predictive
18.The performance of large language models (LLMs) generally improves as more data and parameters are added.
 
False
 
True 정답
19.What action does Google recommend organizations take to ensure that AI is used responsibly?
 
Use a checklist to evaluate responsible AI
 
Follow a top-down approach to increase AI adoption
 
Seek participation from a diverse range of people 정답
 
Focus on being efficient
20.Most organizations have the capability to train large language models (LLMs) from scratch.
 
True
 
False 정답

 

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Responsible AI: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Large Language Models: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Generative AI: Quiz  (0) 2023.10.10
1. Organizations are developing their own AI principles that reflect their mission and values. What are the common themes among these principles?
(1) A consistent set of ideas about transparency, fairness, accountability, and privacy. 정답
(2) A consistent set of ideas about transparency, fairness, and diversity.
(3) A consistent set of ideas about fairness, accountability, and inclusion.
(4) A consistent set of ideas about transparency, fairness, and equity.
 
 
 
2. Which of the below is one of Google’s 7 AI principles?
(1)AI should gather or use information for surveillance.
(2)AI should create unfair bias.
(3)AI should uphold high standards of operational excellence.
(4)AI should uphold high standards of scientific excellence. 정답
 
3. Why is responsible AI practice important to an organization?
(1) Responsible AI practice can help build trust with customers and stakeholders. 정답
(2) Responsible AI practice can help drive revenue.
(3) Responsible AI practice can improve communication efficiency.
(4) Responsible AI practice can help improve operational efficiency.
 
4. Which of these is correct with regard to applying responsible AI practices?
(1) Decisions made at all stages in a project make an impact on responsible AI. 정답
(2) Only decisions made by the project owner at any stage in a project make an impact on responsible AI.
(3) Decisions made at an early stage in a project do not make an impact on responsible AI.
(4) Decisions made at a late stage in a project do not make an impact on responsible AI.

 

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Generative AI Fundamentals Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Large Language Models: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Generative AI: Quiz  (0) 2023.10.10

1. What are some of the applications of LLMs? LLM의 응용 분야에는 어떤 것들이 있습니까?

(1) LLMs can be used for many tasks, including:1) Writing2) Translating3) Coding4) Answering questions5) Summarizing text6) Generating non-creative discrete probabilities, classes, and predictions.

LLM은 다음을 포함하는 많은 작업에 사용될 수 있다:1) 쓰기 2) 번역 3) 코딩 4) 질문에 대한 답 5) 텍스트 요약 6) 비창조 이산 확률, 클래스 및 예측 생성

(2) LLMs can be used for many tasks, including:1) Writing2) Translating3) Coding4) Answering questions5) Summarizing text6) Generating non-creative discrete probabilities
LLM은 다음을 포함하는 많은 작업에 사용될 수 있다:1) 쓰기 2) 번역 3) 코딩 4) 질문에 대한 답 5) 텍스트 요약 6) 비창조 이산 확률 생성


(3) LLMs can be used for many tasks, including:1) Writing2) Translating3) Coding4) Answering questions5) Summarizing text6) Generating non-creative discrete classes

LLM은 다음을 포함하는 많은 작업에 사용될 수 있다:1) 쓰기 2) 번역 3) 코딩 4) 질문에 대한 답 5) 텍스트 요약 6) 비창조 불연속 클래스 생성

(4) LLMs can be used for many tasks, including: 1) Writing2) Translating3) Coding4) Answering questions5) Summarizing text6) Generating non-creative discrete predictions

LLM은 다음을 포함하는 많은 작업에 사용될 수 있다:1) 쓰기 2) 번역 3) 코딩 4) 질문에 대한 답 5) 텍스트 요약 6) 비창조 불연속 예측 생성

(5) LLMs can be used for many tasks, including:1) Writing2) Translating3) Coding4) Answering questions5) Summarizing text6) Generating creativecontent

LLM은 다음을 포함하는 많은 작업에 사용될 수 있다:1) 쓰기 2) 번역 3) 코딩 4) 질문에 대한 답 5) 텍스트 요약 6) 창의적 콘텐츠 생성

 

정답은 (5)번

2. What are some of the challenges of using LLMs? Select three options.
LLM을 사용할 때 어떤 어려움이 있습니까? 세 가지 옵션을 고르시오.
(1)After being developed, they only change when they are fed new data.
개발된 후, 그들은 새로운 데이터를 공급받을 때만 변화한다.
(2)They can be biased.
그들은 편향적일 수 있다.
(3)They can be used to generate harmful content.
그들은 해로운 내용을 생성하는 데 사용될 수 있다.
(4)They can be expensive to train.
그들은 훈련하는 데 많은 돈이 들 수 있다.

정답은 (2),(3),(4)번

 

3. What are some of the benefits of using large language models (LLMs)?

(1) LLMs have a number of benefits, including:1) They can generate human-quality text.2) They can be used for many tasks, such as text summarization and code generation.3) They can be trained on massive datasets of text, images, and code.4) They are constantly improving.
LLM은 다음과 같은 여러 가지 이점이 있습니다: 1) 그들은 사람 수준의 텍스트를 생성할 수 있습니다. 2) 텍스트 요약 및 코드 생성과 같은 많은 작업에 사용될 수 있습니다. 3) 그들은 텍스트, 이미지, 코드의 대규모 데이터 세트에서 훈련될 수 있습니다. 4) 그들은 계속해서 개선되고 있습니다.
 
(2) LLMs have a number of benefits, including:1) They can generate non-probabilities and human-quality text.2) They can be used for many tasks, such as text summarization and code generation.3) They can be trained on massive datasets of text, image, and code.4) They are constantly improving.
 
(3) LLMs have many benefits, including: 1) They can generate probabilities and human-quality text.2) They can be used for many tasks, such as text summarization and code generation.3) They can be trained on massive datasets of text and code.4) They are constantly beingimproved.
 
(4) LLMs have many benefits, including: 1) They can generate discrete classes and human-quality text.2) They can be used for many tasks, such as text summarization and code generation.3) They can be trained on massive datasets of text and code.4) They are constantly improving.
 
(5) LLMs have many benefits, including: 1) They can generate human-quality text. 2) They can be used for a variety of tasks.3) They can be trained on massive datasets of text and code. 4) They are constantlyimproved.
 
정답은 (4)번
 
 
4. What are large language models (LLMs)?:
 
(1) Generative AI is a type of artificial intelligence (AI) that can create new content, such as discrete numbers, classes, and probabilities. It does this by learning from existing data and then using that knowledge to generate new and unique outputs.
 
(2)An LLM is a type of artificial intelligence (AI) that can generate human-quality text. LLMs are trained on massive datasets of text and code, and they can be used for many tasks, such as writing, translating, and coding.
LLM은 사람 수준의 텍스트를 생성할 수 있는 인공지능(AI) 유형입니다. LLM은 텍스트와 코드의 대규모 데이터 세트에서 학습하며, 쓰기, 번역, 코딩 등 많은 작업에 사용될 수 있습니다.
 
(3)Generative AI is a type of artificial intelligence (AI) that only can create new content, such as text, images, audio, and video by learning from new data and then using that knowledge to predict a discrete, supervised learning output.
 
(4) Generative AI is a type of artificial intelligence (AI) that only can create new content, such as text, images, audio, and video by learning from new data and then using that knowledge to predict a classification output.
 
정답은 (2)번

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Generative AI Fundamentals Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Responsible AI: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Generative AI: Quiz  (0) 2023.10.10
1. What is Generative AI?: 생성형 AI란 무엇인가?
 
(1) Generative AI is a type of artificial intelligence (AI) that can create new content, such as discrete numbers, classes, and probabilities. It does this by learning from existing data and then using that knowledge to generate new and unique outputs.
생성 인공지능은 이산형 숫자, 클래스, 확률과 같은 새로운 콘텐츠를 만들 수 있는 인공지능(AI)의 한 종류, 기존의 데이터로부터 학습하고 그 지식을 사용하여 새롭고 독특한 산출물을 생성한다.
(2) Generative AI is a type of artificial intelligence (AI) that can create new content, such as text, images, audio, and video. It does this by learning from existing data and then using that knowledge to generate new and unique outputs.
생성 인공지능은 텍스트, 이미지, 오디오, 비디오와 같은 새로운 콘텐츠를 만들 수 있는 인공지능 (AI)의 한 종류로, 기존의 데이터로부터 학습하고 그 지식을 사용하여 새롭고 독특한 산출물을 생성한다.
(3) Generative AI is a type of artificial intelligence (AI) that can only create new content, such as text, images, audio, and video by learning from new data and then using that knowledge to predict a classification output.
생성형 AI는 새로운 데이터를 학습한 후 그 지식을 활용해 분류 아웃풋을 예측함으로써 텍스트, 이미지, 오디오, 비디오 등 새로운 콘텐츠만 생성할 수 있는 인공지능(AI)의 한 종류이다.
(4) Generative AI is a type of artificial intelligence (AI) that can only create new content, such as text, images, audio, and video by learning from new data and then using that knowledge to predict a discrete, supervised learning output.
생성 AI는 인공지능(AI)의 한 종류로, 새로운 데이터를 통해 학습한 후 그 지식을 활용해 이산적이고 감독된 학습 결과를 예측함으로써 텍스트, 이미지, 오디오, 비디오 등 새로운 콘텐츠만 생성할 수 있다.
 
답은 (2)번
 
2. Hallucinations are words or phrases that are generated by the model that are often nonsensical or grammatically incorrect. What are some factors that can cause hallucinations? Select three options.
할루시네이션은 종종 무의미하거나 문법적으로 부정확한 모델에 의해 생성되는 단어나 구이다. 할로시네이션을 일으킬 수 있는 요인에는 무엇이 있는가? 세 가지 옵션을 선택하라.
 
(1) The model is trained on noisy or dirty data. 모델이 잡음이 많거나 지저분한 데이터로 학습되었다.
 
(2) The model is not trained on enough data. 모델에 학습된 데이터가 충분하지 않았다.
 
(3) The model is not given enough context. 모델에 충분한 문맥이 주어지지 않았다.
 
(4) The model is trained on too much data. 모델이 너무 많은 데이터로 학습되었다.
답은 (1),(2),(3)번

 

3. What is an example of both a generative AI model and a discriminative AI model?
생성형 AI 모델과 식별형 AI 모델에 대한 설명으로 바르게 된 예시는 무엇인가?
 
(1) A generative AI model does not need to be trained on a dataset of images of cats and then used to generate new images of cats, because the images were already generated by using AI. A discriminative AI model could be trained on a dataset of images of cats and dogs and then used to classify new images as either cats or dogs.
생성 AI 모델은 고양이 이미지 데이터 세트에 대해 훈련된 후 고양이 이미지를 생성하는 데 사용될 필요가 없는데, 이는 이미지가 이미 AI를 사용하여 생성되었기 때문이다. 식별형 AI 모델은 고양이와 개 이미지 데이터 세트에 대해 훈련된 후 새로운 이미지를 고양이 또는 개로 분류하는 데 사용될 수 있다.
 
(2) A generative AI model could be trained on a dataset of images of cats and then used to generate new images of cats. A discriminative AI model could be trained on a dataset of images of cats and dogs and then used to classify new images as either cats or dogs.
생성형 AI 모델은 고양이 이미지 데이터 세트에 대해 훈련된 후 고양이 이미지를 생성하는 데 사용될 수 있다. 식별형 AI 모델은 고양이 및 개 이미지 데이터 세트에 대해 훈련된 후 새로운 이미지를 고양이 또는 개로 분류하는 데 사용될 수 있다.
 
(3) A generative AI model could be trained on a dataset of images of cats and then used to classify new images of cats. A discriminative AI model could be trained on a dataset of images of cats and dogs and then used to predict new images as either cats or dogs.
생성형 AI 모델은 고양이 이미지 데이터 세트에 대해 훈련된 후 고양이의 새로운 이미지를 분류하는 데 사용될 수 있다. 식별형 AI 모델은 고양이와 개의 이미지 데이터 세트에 대해 훈련된 후 고양이 또는 개로서 새로운 이미지를 예측하는 데 사용될 수 있다.
 
(4) A generative AI model could be trained on a dataset of images of cats and then used to cluster images of cats. A discriminative AI model could be trained on a dataset of images of cats and dogs and then used to predict as either cats or dogs.
생성형 AI 모델은 고양이 이미지 데이터 세트에 대해 훈련된 후 고양이 이미지를 클러스터링하는 데 사용될 수 있다. 식별형 AI 모델은 고양이 및 개 이미지 데이터 세트에 대해 훈련된 후 고양이 또는 개 중 하나로 예측하는 데 사용될 수 있다.
 
답은 (2)번
 
4. What is a prompt? 프롬프트란 무엇인가?
(1) A prompt is a short piece of text that is given to the large language model as input, and it can be used to control the output of the model in many ways.
프롬프트는 큰 언어 모델에 입력으로 제공되는 짧은 텍스트로서, 모델의 출력을 제어하기 위해 여러 가지 방법으로 사용될 수 있다.
 
(2) A prompt is a short piece of text that is given to the small language model (SLM) as input, and it can be used to control the output of the model in many ways.
프롬프트는 작은 언어 모델(small language model, SLM)에 입력으로 제공되는 짧은 텍스트로서, 다양한 방법으로 모델의 출력을 제어하기 위해 사용될 수 있다.
 
(3) A prompt is a short piece of text that is given to the large language model as input, and it can be used to control the input of the model in many ways.
프롬프트는 큰 언어 모델에 입력으로 제공되는 짧은 텍스트로서, 모델의 입력을 제어하기 위해 여러 가지 방법으로 사용될 수 있다.
 
(4) A prompt is a short piece of code that is given to the large language model as input, and it can be used to control the output of the model in many ways.
프롬프트는 입력으로 큰 언어 모델에 부여되는 짧은 코드 조각으로, 여러 가지 방법으로 모델의 출력을 제어하는 데 사용될 수 있다.
 
(5) A prompt is a long piece of text that is given to the large language model as input, and it cannot be used to control the output of the model.
프롬프트는 큰 언어 모델에 입력으로 제공되는 긴 텍스트이며, 모델의 출력을 제어하기 위해 사용될 수 없다.
 
답은 (1)번
 
5. What are foundation models in Generative AI? Generative AI의 기초 모델은 무엇인가?
(1) A foundation model is a large AI model post-trained on a vast quantity of data that was "designed to be adapted” (or fine-tuned) to a wide range of downstream tasks, such as sentiment analysis, image captioning, and object recognition. 기초 모델은 감정 분석, 이미지 캡션 및 객체 인식과 같은 광범위한 다운스트림 작업에 "적응되도록 설계"(또는 미세 조정)된 방대한 양의 데이터에 대해 사후 훈련된 대형 AI 모델이다.
(2) A foundation model is a large AI model both post and pre-trained on a vast quantity of data that was "designed to be adapted” (or fine-tuned) to a wide range of downstream tasks, such as sentiment analysis, image captioning, and object recognition.
기초 모델은 감정 분석, 이미지 캡션 및 객체 인식과 같은 광범위한 다운스트림 작업에 "적응되도록 설계"(또는 미세 조정)된 방대한 양의 데이터를 게시 및 사전 교육한 대형 AI 모델이다.
(3) A foundation model is a large AI model pretrained on a vast quantity of data that was "designed to be adapted” (or fine-tuned) to a wide range of upstream tasks, such as sentiment analysis, image captioning, and object recognition.
기초 모델은 감정 분석, 이미지 캡션 및 객체 인식과 같은 광범위한 업스트림 작업에 "적응되도록 설계"(또는 미세 조정)된 방대한 양의 데이터를 사전에 훈련한 대형 AI 모델이다.
(4) A foundation model is a small AI model pretrained on a small quantity of data that was "designed to be adapted” (or fine-tuned) to a wide range of downstream tasks, such as sentiment analysis, image captioning, and object recognition.
기초 모델은 감정 분석, 이미지 캡션 및 객체 인식과 같은 광범위한 다운스트림 작업에 "적응되도록 설계"(또는 미세 조정)된 소량의 데이터를 사전 교육한 소규모 AI 모델이다.
(5) A foundation model is a large AI model pretrained on a vast quantity of data that was "designed to be adapted” (or fine-tuned) to a wide range of downstream tasks, such as sentiment analysis, image captioning, and object recognition.
기초 모델은 감정 분석, 이미지 캡션 및 객체 인식과 같은 광범위한 다운스트림 작업에 "적응되도록 설계"(또는 미세 조정)된 방대한 양의 데이터를 사전에 훈련한 대형 AI 모델이다.

답은 (5)번

'인공지능(ML) > 생성형 AI' 카테고리의 다른 글

[Google ML Study Jam] 생성형 AI(Gen AI) - Encoder-Decoder Architecture: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Image Generation: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Generative AI Fundamentals Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Responsible AI: Quiz  (0) 2023.10.11
[Google ML Study Jam] 생성형 AI(Gen AI) - Introduction to Large Language Models: Quiz  (0) 2023.10.11

+ Recent posts

티스토리툴바