VAE (Variational Autoencoder)
Definition
A generative model that learns a compressed latent representation of data and can generate new samples by sampling from this learned distribution.
Variational Autoencoders combine ideas from deep learning and Bayesian inference. An encoder network maps input data to a probability distribution in latent space, and a decoder network reconstructs data from samples drawn from that distribution. The model is trained to minimize both reconstruction error and the divergence between the learned distribution and a prior (typically Gaussian). VAEs produce smooth, continuous latent spaces that allow meaningful interpolation between data points. While VAE outputs tend to be blurrier than those from GANs or diffusion models, they are essential components in modern generative systems. Stable Diffusion, for example, uses a VAE to compress images into a latent space where the diffusion process operates more efficiently.
Related Terms
Diffusion Model
A generative model that learns to create data by gradually denoising a random noise signal, reversin...
GAN (Generative Adversarial Network)
A generative model architecture consisting of two neural networks — a generator and a discriminator ...
Generative AI
AI systems that can create new content such as text, images, audio, video, and code, rather than sim...
Embedding
A learned dense vector representation that maps discrete data like words, tokens, or items into cont...