Skip to content

Mean-Recall Approximation for the Leitner Queue Network

Theory · All theories
Evidence · 3 claims (2 for, 1 mixed) · 1 study (1 causal), q1 · 0 of 1 report an effect size · 3 claims rest on one study

Description

The mean-recall approximation is the article's heuristic that treats each Leitner deck as an M/M/1 queue, replacing each item's true delay with an exponential draw so recall probability becomes a function of deck service and input rates. Under it, "the problem of choosing an optimal review schedule reduces to a low-dimensional deter- ministic optimization problem" that a nonlinear solver can solve.

Design Implications

Context

Requirements

  • Arrival and input rates below each deck's service rate so that queues are ergodic.
  • Flow-balance equations linking deck input rates to recall probabilities.
  • A known item difficulty θ and review budget U to solve the static planning problem.

Constraints

  • The approximation is essentially a heuristic; more rigorous approximations are an open topic for future work.
  • Simulations suggest it matches the clocked-delay model closely only for small arrival rates.
  • If a deck's input rate is not below its service rate, that deck's size and review delays grow unbounded.

Target Learners

  • Learners using spaced repetition flashcard systems in the long-term learning regime

Target Learning Objectives

  • Optimizing flashcard review schedules to maximize the rate of learning

Claims

Examples

-

Key Sources

  • Reddy, S., Labutov, I., Banerjee, S., & Joachims, T. (2016). Unbounded Human Learning: Optimal Scheduling for Spaced Repetition. KDD ’16, San Francisco, CA, USA. https://doi.org/10.1145/2939672.2939850
Page metadata
Field Value
type theory
title Mean-Recall Approximation for the Leitner Queue Network
description The mean-recall approximation is the article's heuristic that treats each Leitner deck as an M/M/1 queue, replacing each item's true delay with an exponential draw so recall probability becomes a function of deck serv...
status review
canonical True
generated by claude/unspecified on 2026-09-24
trust unverified — no one has confirmed this page's content
sources 1 — reddy-2016

Frontmatter is the page's machine-readable record — see the Schema & Guide.