Diffusion Model
Generative models that learn to create data by reversing a gradual noising process.
In-depth explanation
Diffusion models add noise to data in small steps, then learn to reverse this process, generating data from pure noise. They've achieved state-of-the-art image generation quality, surpassing GANs. Stable Diffusion, DALL-E, and Midjourney use diffusion. The approach is more stable to train than GANs and offers better sample diversity.
Examples
More in Generative AI
Generative Adversarial Network (GAN)
Two neural networks competing against each other to generate realistic synthetic data.
GPT
Generative Pre-trained Transformer, a family of large language models trained to generate text.
Hallucination
When AI models generate plausible-sounding but factually incorrect or fabricated information.
Large Language Model (LLM)
AI models trained on vast text data that can generate and understand human-like text.
Prompt Engineering
The practice of crafting effective inputs to get desired outputs from AI models.
RAG (Retrieval-Augmented Generation)
Combining retrieval systems with language models to generate responses grounded in external knowledge.
Master Diffusion Model.
Learn how to apply this concept with hands-on projects in our comprehensive AI programs.