AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Relaxed Marginal Consistency for Differentially Private Query Answering

Ryan McKenna, Siddhant Pradhan, Daniel Sheldon, Gerome Miklau

NeurIPS 2021 · Federated learning and privacy › Differential privacy · approach: Privacy accounting and mechanisms · 1 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
replace exact inference with a saturated region graphProx-PGM · Synthetic 100-dimensional data, random and greedy 3-way marginalsmeasured cliques handled: about 50; 1000 -> 10000Improved
use region-graph reconstructionHDMM · fire, 455 random 3-way marginalsRMSE: 3 \times reductionImproved
use a saturated region graph instead of exact inference (ablation)Prox-PGM · Synthetic 8-dimensional data, in-model 3-way marginalsL1 errorHurt
use a factor graph instead of exact inference (ablation)Prox-PGM · Synthetic 8-dimensional data, in-model 3-way marginalsL1 errorHurt
use a factor graph instead of a saturated region graph (ablation)Prox-PGM · Synthetic 8-dimensional data, out-of-model marginalsL1 errorHurt
use a saturated region graph instead of exact inference (ablation)Prox-PGM · Synthetic 8-dimensional data, out-of-model marginalsL1 errorHurt
use APPGM rather than Private-PGMMWEM · fire, all 2-way marginalserror at a fixed number of roundsHurt
run MWEM with APPGM to completion rather than stop at Private-PGM's minimum-error pointMWEM · Additional MWEM experimentserror: 3× as many rounds; 3× as much privacy budgetImproved
use APPGM to reconstruct FEM answersFEM · AdulterrorImproved
use an over-saturated region graph to include out-of-model marginals (ablation)APPGM · Out-of-model marginal estimationestimate qualityImproved
use an over-saturated region graph to include out-of-model marginals (ablation)APPGM · Out-of-model marginal estimationruntimeHurt
warm start Convex-GBP messages and run one inner iteration per Prox-PGM iteration (ablation)APPGM · Practical optimizationobjective progressImproved

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