Glossary term
Glossary term
Architecture
A type of autoencoder that leverages the discrepancy between inputs and outputs to generate modified versions of the inputs. Variational autoencoders are useful for generative AI.
VAEs are based on variational inference: a technique for estimating the parameters of a probability model.
Created for this library
A research team uses a variational autoencoder to learn compact latent representations of customer behavior for downstream segmentation.
A medical imaging team uses a variational autoencoder to compress images and detect anomalies with unusual reconstruction error.
A fraud team uses a variational autoencoder on transaction patterns to flag transactions with unusual latent representations.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License