Basic part 2

Question 1

prompt

Answer 1

short piece of text that is used to provide input or instructions to a generative AI system.

Question 2

Models

Answer 2

Models are at the core of generative AI and are built with neural networks, system resources, data, and prompts, all working together.

Question 3

diffusion models

Answer 3

Generative models focused on creating new images by adding and removing noise to input images repeatedly over a series of iterations. The final image ends up looking similar to the images used as part of the training dataset.

Question 4

embeddings

Answer 4

Language models are trained to understand and generate natural language by representing words and sentences with encoded mathematical vectors

Question 5

Large Language Model

Answer 5

A type of artificial intelligence system that is trained on massive amounts of text data to generate human-like responses.

Question 6

model training

Answer 6

The process of teaching a generative AI model how to create new content by showing the model many examples of the kind of content you want it to create, like images, text, or music.

Question 7

neural network

Answer 7

a neural network is a mathematical algorithms that learns by finding statistical patterns in data. It consists of layers of artificial neurons - the first layer receives the input data, with subsequent layers processing additional information. The last layer in the network outputs the results.

Question 8

parameters

Answer 8

Numerical values that define the overall characteristics (i.e., structure and behavior) of a large language model.

Question 9

positional encoding

Answer 9

a technique used to identify the order and position of words

Question 10

self-attention

Answer 10

A mechanism used in generative AI models to help the model understand relationships between words and sentences.

Question 11

transformer

Answer 11

Commonly used in natural language processing, transformers are a type of deep learning model designed to understand the contextual relationship between sequences of text within a sentence or paragraph.

Question 12

fine tuning

Answer 12

The process of taking an existing large language model and training it further on a smaller dataset that is specific to a certain task

Question 13

hallucination

Answer 13

refers to when a generative AI model creates content that is not actually present in the input data it was trained on.

Question 14

sentiment analysis

Answer 14

the use of natural language processing (NLP) techniques to determine the emotional tone or attitude expressed in a piece of text.

Question 15

text summarization

Answer 15

the process of taking a long piece of text and generating a shorter version that captures the key points. Large language models can be trained to perform text summarization automatically. A large language model is shown many examples of texts paired with human-written summaries during training

Question 16

tokens

Answer 16

Tokens are building blocks used by the model to understand the meaning of the text and generate a response.

Question 17

weights

Answer 17

numeric values that represent the strength of the connections in a neural network, which are tuned by training and allow the model to make intelligent predictions.

Question 18

Intelligent Document Processing

Answer 18

used to automatically extract structured data from various types of documents, such as invoices, contracts, and forms?

Question 19

Regression

Answer 19

Predicting a numerical value

Question 20

Classification

Answer 20

Predicting a label

Question 21

Ranking

Answer 21

Ordering items to find the most relevant

Question 22

Recommendation

Answer 22

Finding relevant items based on past behavior

Question 23

Clustering

Answer 23

Finding patterns in examples

Question 24

Anomaly detection

Answer 24

Finding outliers from examples