Variational autoencoder (VAE) DL
noun phrase
Definition: A generative autoencoder model that learns compressed representations of training data as probability distributions in latent space and uses them to generate new samples by decoding variations of those learned representations [IBM Think].
Example in context: “Subsequently, it employs a generative neural network, specifically a variational autoencoder, to create novel molecular structures that are computationally predicted to t arget disease-associated genes and biological processes.” [Ghebrehiwet et al. 2024]
Synonyms : VAE
Related terms: autoencoder, latent space, generative model, GAN