AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Algorithms for bounding contribution for histogram estimation under user-level privacy

Yuhan Liu, Ananda Theertha Suresh, Wennan Zhu, Peter Kairouz, Marco Gruteser

ICML 2023 · Federated learning and privacy › Differential privacy · approach: Local-DP and private estimation · 0 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
Select the contribution threshold privately using the dataset-specific objective instead of the median thresholdUnbounded-domain histogram estimation · All reported synthetic and Sentiment140 settingsrelative loss: i.i.d., d=50: 0.0015 vs 0.0397; non-i.i.d., d=50: 0.0124 vs 0.6957; Sentiment140, d=1000: 0.1483 vs 0.5912Improved
Select the contribution threshold privately using the dataset-specific objective instead of the 90% quantile thresholdUnbounded-domain histogram estimation · All reported synthetic and Sentiment140 settingsrelative loss: i.i.d., d=50: 0.0015 vs 0.0049; non-i.i.d., d=50: 0.0124 vs 0.1645; Sentiment140, d=1000: 0.1483 vs 0.2699Improved
Use private threshold selection rather than the best threshold chosen non-privatelyUnbounded-domain histogram estimation · i.i.d. synthetic, d=50, d=100, and d=200relative loss: 0.001 -> 0.0015; 0.0026 -> 0.0026; 0.0048 -> 0.0048No clear effect
Use private threshold selection rather than the best threshold chosen non-privatelyUnbounded-domain histogram estimation · non-i.i.d. synthetic, d=50, d=100, and d=200relative loss: 0.0094 -> 0.0124; 0.0175 -> 0.0224; 0.0461 -> 0.0462No clear effect
Use private threshold selection rather than the best threshold chosen non-privatelyUnbounded-domain histogram estimation · Sentiment140, d=100 and d=1000relative loss: 0.0076 -> 0.0285; 0.1250 -> 0.1483Mixed
Estimate the bounded-domain Gaussian clipping threshold with DP-SGD rather than the DP-M-quantile heuristicBounded-domain histogram estimation · Sentiment140 and SNAPestimation errorImproved
Use DP-M-quantile to set the clipping thresholdBounded-domain histogram estimation · Sentiment140 and SNAPestimation errorHurt
Apply Poisson-based post-processing to the clipped count estimate to reduce clipping biasCount estimation under non-i.i.d. Poisson user counts · Count estimationsquared error of the mean estimateImproved

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