Sign inSign in
ConceptAI & ML

Autoencoder

A neural network architecture that maps input data to a compressed lower-dimensional representation and then decodes it back to reconstruct the original data.
as Phillip Isola defines itMITThe wording closest to all 5 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.

Neural networks5Dimensionality reduction1Loss function1Loss function optimization1

Opens up

Concepts that name this one as a prerequisite.

Lecturers

4

Each in their own terms, from the easiest level.

  • Phillip Isola

    MIT · Undergraduate

    A model that maps data to a low-dimensional representation and then decodes it back to reconstruct the original data.
    • 4:27Generative Models: Representation Learning Meets Generative Modeling
  • Fei-Fei Li, Justin Johnson & Serena Yeung

    Stanford University · Undergraduate

    An unsupervised learning model that learns a lower-dimensional feature representation from unlabeled training data.
  • Holger Lyre

    University of Magdeburg · Graduate

    An encoder-decoder structure with a bottleneck that learns efficient codings of unlabeled data.
    • 21:37Understanding AI: Semantic Grounding in Large Language Models
  • John M. Aiken

    UT Austin · Research

    A neural network architecture that takes high-dimensional input features, compresses them down into a low-dimensional latent space representation, and then reconstructs the features back out.
    • 16:40Exploring System Dynamics in the Natural World with AI

Lectures that assume it

4

Connections

16

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

4 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.