Generative AI
Diffusion Models
A class of generative models that create new data by learning to reverse a process that gradually adds noise to a dataset.
Explanation
Diffusion models are a type of generative artificial intelligence that works through a two-step process: forward diffusion and reverse diffusion. In the forward phase, Gaussian noise is incrementally added to input data (such as an image) until it becomes unrecognizable. In the reverse phase, a neural network is trained to remove that noise step-by-step to recover the original data. Once trained, the model can generate entirely new samples by starting with pure noise and applying the learned denoising process. These models have largely surpassed Generative Adversarial Networks (GANs) in quality and stability, forming the backbone of modern image generation tools like Stable Diffusion, Midjourney, and DALL-E.