Sign inSign in
ConceptAI & ML

Value iteration

An algorithm that uses the Bellman equation as an iterative update to find the optimal Q-function.
as Fei-Fei Li, Justin Johnson & Serena Yeung defines itStanford UniversityThe wording closest to all 9 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

6

Each in their own terms, from the easiest level.

  • Russ Tedrake

    MIT · Undergraduate

    An iterative algorithm based on the Bellman equation to find the optimal cost-to-go and policy over an infinite horizon.
    • 48:39Dynamic Programming I
    • 20:08Dynamic Programming III
    • 40:51Stochastic Control
  • Pieter Abbeel

    UC Berkeley · Undergraduate

    An algorithm that computes optimal values by starting at the bottom of the search tree and working upward iteratively.
    • 1:07:29Markov Decision Processes (MDPs) Part 1/2
    • 25:42Markov Decision Processes (MDPs) Part 2/2
  • David Silver

    Google DeepMind · Undergraduate

    An iterative algorithm that applies the Bellman optimality equation as an update rule to find the optimal value function by working backwards from the goal.
    • 1:04:53Planning by Dynamic Programming — Policy & Value Iteration
  • Pieter Abbeel & Daniel Klein

    UC Berkeley · Undergraduate

    A backward pass over a trajectory to find the optimal sequence of feedback controllers.
    • 53:43Robotics, Language, and Vision
  • Sergey Levine

    UC Berkeley · Graduate

    An algorithm that short-circuits policy iteration by updating Q-values and directly taking the maximum over actions to compute the value function.
    • 14:21Foundational concepts in deep reinforcement learning

Lectures that assume it

6

Connections

25

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

6 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.