Home▸Data Science▸Base-Rate Fairness Impossibility Lab (2D)

Base-Rate Fairness Impossibility Lab (2D)

Interactive 2D demonstration of the fairness impossibility result: when two groups have different base rates, a classifier that is equally accurate for both (equal FPR/TPR) cannot also be equally calibrated (equal PPV) — drag to pan, scroll to zoom the score strip.

Data Science2DModerate60 FPS📱 Mobile-adapted⇄ 3D version

A machine learning classifier scores two demographic groups using exactly the same underlying risk model — yet the groups have different true base rates for the outcome being predicted. This simulator renders both populations as 2D score strips (drag to pan, scroll to zoom) and lets you set each group's base rate and the decision threshold(s), then watches equalized-odds metrics (FPR, TPR) and the calibration metric (PPV) update live. The result reproduces a real mathematical fact known as the fairness impossibility theorem: when base rates differ, equal accuracy across groups and equal calibration across groups cannot both hold at once, except in the trivial case where the base rates already match.

⚙ Under the hood

Drag-and-zoom 2D score strips show why a classifier that is equally accurate for two groups (equal FPR/TPR) cannot also be equally calibrated (equal PPV) when their base rates differ — Chouldechova's fairness impossibility theorem, live.

fairnessmachine-learningclassificationstatisticsbiasdata-sciencealgorithmic-fairness

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install