Sign inSign in
ConceptAI & ML

Variational autoencoder

A variational autoencoder is an amortized variational inference model that models an input xx using a latent variable zz through an encoder and a decoder.
as Sergey Levine defines itUC BerkeleyThe wording closest to all 4 definitions
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Opens up

Concepts that name this one as a prerequisite.

Lecturers

3

Each in their own terms, from the easiest level.

  • Fei-Fei Li, Justin Johnson & Serena Yeung

    Stanford University · Undergraduate

    A probabilistic version of an autoencoder that allows sampling from the model to generate new data.
  • Phillip Isola

    MIT · Undergraduate

    A model that combines an encoder and a decoder to learn a mapping between data and a latent space that is easy to sample from.
    • 4:21Generative Models: Representation Learning Meets Generative Modeling
  • Sergey Levine

    UC Berkeley · Graduate

    A generative model that has some latent variable representation of an image or state.
    • 3:02Exploration of advanced topics in deep reinforcement learning

Lectures that assume it

3

Connections

17

The lectures that define it or assume it, and the concepts next to it.

3 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.