Cognitive Scaffold

Preparing your thinking workspace

arrow_back_ios_new
MENTAL MODEL · M10732

Shapley Value Decomposition

Shapley Value Decomposition
DecideHigh supportGame Theory
Included
account_tree

Updated 2026-08-13

Loading revision record…

INTRODUCTION

English translation pending.

CORE DEFINITION

The Shapley value is a solution concept in cooperative game theory that allocates the total payoff of a coalition by computing each player's average marginal contribution across every possible order in which the coalition could form. The core proposition is that this allocation satisfies fairness axioms including efficiency, symmetry, and the treatment of players who add nothing, so it provides a principled way to attribute a jointly produced result to its contributors. The key qualification is that the decomposition depends on a defined baseline, since the value of a factor is measured relative to the empty coalition or another reference point.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

- Baseline definition: state the empty set or reference case against which contributions are measured. - Contribution averaging: compute each factor's marginal contribution across many coalition orders. - Interaction check: verify that interaction effects are handled rather than assumed away.

anchor

Anchor fast decisions

Marginal contribution depends on what is already present, so a factor's apparent importance varies with the order of addition. Averaging over all orders removes that order dependence and yields a single allocation that no participant can improve by reordering. This is why the method attributes joint results more fairly than simple additive accounting, which ignores interaction entirely.

MINIMUM ACTION

In progress 0/1

Practice this model in one real situation:

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/Shapley_valueZH · Explicit
    verified

RELATED MODELS