Generative AI Introduction and Applications Glossary

A comprehensive glossary of terms related to Generative AI, machine learning, and deep learning architectures.

What is data augmentation?

A technique commonly used in machine learning and deep learning to increase the diversity and amount of training data.

What is the definition of deep learning?

A subset of machine learning that focuses on training computers to perform tasks by learning from data through the use of artificial neural networks.

How does a diffusion model generate high-quality samples?

It is trained by gradually adding noise to an image and then learning to remove the noise in a process called diffusion.

What is the primary function of discriminative AI?

It is a type of artificial intelligence that distinguishes between different classes of data.

What are discriminative AI models generally used for?

They identify and classify based on patterns observed in training data and are used in prediction and classification tasks.

What are foundation models?

AI models with broad capabilities that can be adapted to create more specialized models or tools for specific use cases.

What are the two neural networks in a Generative adversarial network (GAN)?

A generator, which is trained to create samples like text and images, and a discriminator, which tries to distinguish whether the sample is real or fake.

How is Generative AI defined?

A type of artificial intelligence that can create new content, such as text, images, audio, and video.

How do generative AI models produce original content?

They analyze the patterns and underlying structure of training data to generate entirely new content, typically used for automated content creation and interactive communication.

What is a Generative pre-trained transformer (GPT)?

A family of large language models developed by OpenAI that use the Transformer architecture and are pre-trained on large amounts of text data to understand and generate human-like language.

What tasks can Large language models (LLMs) perform?

They can perform language-related tasks, including text generation, translation, summarization, sentiment analysis, and more.

What is the core focus of machine learning?

Creating algorithms and models that enable computers to learn and make predictions or decisions from training data.

What is Multimodal AI?

AI programs with the capability to process a variety of inputs, including text, images, audio, and video, and convert these inputs into different formats.

What does Natural language processing (NLP) enable computers to do?

It enables computers to understand, manipulate, and generate human language (natural language).

What are neural networks?

Computational models inspired by the structure and functioning of the human brain, serving as a fundamental component of deep learning and artificial intelligence.

In generative AI, what is a prompt?

Inputs—such as text instructions, questions, images, or audio—provided to a generative AI model to guide and generate new content.

What is the role of training data?

It consists of large datasets, often including examples, used to teach a machine learning model.

What is the defining mechanism of Transformers?

A deep learning architecture designed to process sequential data using self-attention mechanisms, forming the backbone of modern large language models.

What is a Variational autoencoder (VAE)?

A type of generative model and neural network designed to learn efficient representation of input data by encoding it into a smaller space and decoding back to the original space.