Latent space is the compressed, abstract representation of data learned by a neural network — a lower-dimensional space where meaningful features are encoded. In a well-trained latent space, similar inputs are close together and interpolation between points produces semantically meaningful transitions. Autoencoders and VAEs learn latent spaces explicitly; GANs and diffusion models sample from them to generate new data. Latent spaces enable powerful capabilities like finding similar images, morphing between concepts, and anomaly detection.
Deep Learning
Latent Space
Related terms
Learn to use these concepts in practice.
Join the 212AY Academy →