Sign inSign in
ConceptAI & ML

N-step returns

Returns accumulated over N steps instead of a single step to compute targets for Q-learning.
as Sergey Levine defines itUC BerkeleyThe wording closest to all 3 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

2

Each in their own terms, from the easiest level.

  • Sergey Levine

    UC Berkeley · Undergraduate

    An n-step backup method used in reinforcement learning to compute returns over multiple time steps before bootstrapping.
    • 4:19Exploration of advanced topics in deep reinforcement learning
  • David Silver

    Google DeepMind · Undergraduate

    The return calculated from nn steps of actual rewards plus the discounted value estimate at state St+nS_{t+n}.
    • 1:17:30Model-Free Prediction — Monte Carlo, TD(0) & TD(λ)

Lectures that assume it

1

Connections

11

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.