What specific type of Autoencoder is designed to generate new data by learning the parameters of a probability distribution?
Show answer
Variational Autoencoder
Variational Autoencoders (VAEs) add a probabilistic twist to the traditional autoencoder by constraining the latent space to follow a Gaussian distribution. This allows for sampling from the latent space to generate new, synthetic data points.
Also asked as:
- Which type of autoencoder allows for generating synthetic data by sampling from a latent space constrained to a Gaussian distribution?