Sign inSign in
ConceptAI & ML

Self-supervised learning

A learning paradigm where systems learn from unlabelled data by predicting missing parts of the input.
as Yann LeCun defines itHarvard UniversityThe 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.

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

    Self-supervised learning uses supervised learning machinery to predict raw data or parts of data instead of human-annotated labels.
    • 1:06:36Representation Learning: Reconstruction-Based
  • Steven Feng

    Stanford University · Undergraduate

    A training approach where models learn by corrupting raw data and training to reconstruct the original uncorrupted data to learn general representations.
    • 8:00Overview of Transformers (Stanford CS25: Transformers United V6)
  • Yann LeCun

    Harvard University · Graduate

    • 15:55The Deep Learning Revolution: The Sequel (Turing Award Lecture)
  • Holger Lyre

    University of Magdeburg · Graduate

    Learning models of unlabeled data to extract hidden causes and latent variables in order to sample or make inferences about unseen data.
    • 24:04Understanding AI: Semantic Grounding in Large Language Models

Lectures that assume it

5

Connections

19

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.