Sign inSign in
ConceptAI & ML

Teacher forcing

A training technique for sequence models where the ground-truth output from the previous time step is used as the input for the current time step.
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.

Lecturers

2

Each in their own terms, from the easiest level.

  • Phillip Isola

    MIT · Graduate

    A training method for autoregressive models where each next word prediction is conditioned on the ground-truth preceding word rather than the model's own predictions.
    • 17:29Inference Methods for Deep Learning

Connections

6

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

Concepts4

Lectures2

2 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.