True count Noisy release High-error rod
⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.

Differential Privacy Across Districts: 3D Laplace Mechanism

Real government open-data releases rarely publish just one number — they publish a table: one aggregate statistic per district, program or category, all spent from the same privacy budget. This 3D simulator renders several synthetic districts as paired bars — the true enrollment count in a municipal benefit program next to the noisy count the Laplace mechanism would actually publish — and lets you tune a single shared privacy budget ε across all of them at once. Run hundreds of independent releases per district to compute the real, honestly-sampled mean absolute error and accuracy rate at the current ε, and orbit the scene to see how the same mechanism behaves across districts of very different sizes.