AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

An Uncertainty Principle is a Price of Privacy-Preserving Microdata

John M. Abowd, Robert Ashmead, Ryan Cumings-Menon, Simson L. Garfinkel, Daniel Kifer, Philip Leclerc, William Sexton, Ashley E Simpson, Christine Task, Pavel Zhuravlev

NeurIPS 2021 · Federated learning and privacy › Differential privacy · approach: Privacy accounting and mechanisms · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use ReWeight instead of NNLS postprocessingLaplace mechanism microdata · Sum query, D01squared error: 461.9 -> 108.5Improved
use Sequential fitting instead of NNLSLaplace mechanism microdata · Sum query, D01squared error: 461.9 -> 149.2Improved
use MaxFit instead of OLS (nonnegative microdata)Laplace mechanism · Sum query, D06squared error: 107.2 -> 677.4Hurt
use NNLS instead of OLS (nonnegative microdata)Laplace mechanism · Sum query, D06squared error: 107.2 -> 584.0Hurt
use ReWeight instead of NNLSLaplace mechanism microdata · Sum query, D13squared error: 300.3 -> 108.8Improved
use ReWeight instead of NNLSLaplace mechanism microdata · Identity queries total, D01total squared error: 344.2 -> 159.2Improved
use ReWeight instead of NNLSLaplace mechanism microdata · Identity queries max cell error, D08max squared error: 130.0 -> 168.9Hurt
use ReWeight instead of NNLSLaplace mechanism microdata · Identity queries total, D13total squared error: 2884.1 -> 2870.4No clear effect
use Sequential instead of ReWeightLaplace mechanism microdata · Identity queries total, D13total squared error: 2870.4 -> 2484.2Improved
use ReWeight instead of SequentialLaplace mechanism microdata · Sum query, D01squared error: 149.2 -> 108.5Improved
use ReWeight instead of NNLSLaplace mechanism microdata · Identity queries max cell error, D01max squared error: 147.4 -> 78.4Improved
use ReWeight instead of NNLSLaplace mechanism microdata · Identity queries max cell error, D13max squared error: 119.2 -> 166.1Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy