Definition
Diffusion models work by adding noise to data gradually, then learning to reverse this process to generate new samples.
How It Works: 1. Forward process: Add noise step by step until pure noise 2. Reverse process: Learn to denoise step by step 3. Generation: Start from noise, denoise to create new data
Key Advantages: - High-quality outputs - Training stability - Flexible conditioning
Popular Implementations: - Stable Diffusion - DALL-E 2/3 - Midjourney - Imagen
Applications: - Image generation - Video synthesis - Audio generation - 3D asset creation
Examples
Stable Diffusion generating photorealistic images from text prompts.
Related Terms
Want more AI knowledge?
Get bite-sized AI concepts delivered to your inbox.
Free daily digest. No spam, unsubscribe anytime.