MENTAL MODEL · M9263

Differential Evolution

CreateHigh supportInnovation Methods

Included

Updated 2026-08-15

Loading revision record…

INTRODUCTION

English translation pending.

CORE DEFINITION

Differential evolution is a global optimization algorithm that evolves increasingly good solutions through differential mutation, crossover, and selection over the solution space. Its transferable insight is that the differences between individuals can point the way to improvement, and that idea generalizes to creative thinking: the gap between two existing approaches can supply the direction of the next one rather than blind search.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

- Use the difference: let the gap between existing options suggest the next move. - Combine before selecting: build the candidate by mixing, then keep it only if it wins. - Tune the step: set how far the difference pushes you before you run it.

anchor

Anchor fast decisions

A random mutation has no information about where improvement lies, but the difference vector between two population members is a direction already tested by survival. Adding that difference to a third member sends the candidate along a gradient the population itself has discovered, so the search is informed by its own history rather than blind.

MINIMUM ACTION

In progress 0/1

Practice this model in one real situation:

Initialize a population of candidate solutions spread across the space. 2. For each individual, build a trial vector by differential mutation and crossover. 3. Compare the trial with its parent on the objective. 4. Keep the better of the two into the next generation. 5. Tune the scale factor and the crossover rate until it converges.

Check to track your progress (stored locally)

Learning progress0%

account_treeGenealogyexpand_more

menu_bookReferencesexpand_more

Source support: Explicit

link
en.wikipedia.orghttps://en.wikipedia.org/wiki/Differential_evolutionZH · Explicit
verified

RELATED MODELS