Model Components
Generative Model
Generative models learn the underlying data distribution P(x) and can sample new instances from it. Types include autoregressive models (GPT), variational autoencoders (VAE), generative adversarial networks (GAN), and diffusion models (Stable Diffusion, DALL-E).
Generative models power text generation, image synthesis, music generation, and drug discovery. They are distinct from discriminative models (which learn P(y|x) for classification) in that they can produce novel outputs rather than just categorizing inputs.
Authority Links
Related Terms
Core Concepts
Generative AI
AI systems that produce new content — text, images, audio, video, or code — by learning the statistical distributions of training data and sampling from them, rather than retrieving stored outputs.
Model Components
Generative Adversarial Network (GAN)
Framework training two competing networks—a generator and discriminator—to produce realistic synthetic data.
Model Components
Large Language Model (LLM)
A transformer-based neural network with billions to trillions of parameters, trained on broad text corpora to predict the next token and able to generate, summarize, classify, and reason over natural language.
Model Components
Autoregressive Model
Model that generates each output element by conditioning on all previously generated elements.