Sign inSign in
ConceptAI & ML

Tanh activation function

An activation function that squashes numbers to [−1,1][-1, 1], is zero-centered, but still kills gradients when saturated.
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

2

Each in their own terms, from the easiest level.

  • Sara Beery

    MIT · Graduate

    A pointwise non-linearity bounded between -1 and +1 that outputs values centered at zero.
    • 32:24Introduction to Deep Learning

Lectures that assume it

1

Connections

6

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

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.