AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

An Optimal and Scalable Matrix Mechanism for Noisy Marginals under Convex Loss Functions

Yingtai Xiao, Guanlin He, Danfeng Zhang, Daniel Kifer

NeurIPS 2023 · Learning theory and algorithms › Matrix and tensor methods · approach: Low-rank approximation theory · 2 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 ResidualPlanner instead of HDMM for selection (RMSE objective)HDMM · Synth-nd, d=15selection time: 201.485 ± 13.697 sec -> 0.030 ± 0.017 secImproved
use ResidualPlanner instead of HDMM for selection (RMSE objective)HDMM · Synth-nd, d=10selection time: 0.639 ± 0.059 sec -> 0.009 ± 0.001 secImproved
use ResidualPlanner where HDMM runs out of memoryHDMM · Synth-nd, d=20 to 100selection time: Out of memory -> 9.913 ± 0.246 secImproved
optimize max variance instead of RMSE objective in ResidualPlannerResidualPlanner · Synth-nd, d=100selection time: 9.913 ± 0.246 sec -> 121.224 ± 3.008 secHurt
optimize max variance instead of RMSE objective in ResidualPlannerResidualPlanner · Synth-nd, d=2selection time: 0.001 ± 0.0008 sec -> 0.007 ± 0.001 secNo clear effect
use ResidualPlanner instead of HDMM + PGM for reconstructionHDMM + PGM · Synth-nd, d=15reconstruction time: 91.101 ± 7.621 sec -> 0.413 ± 0.006 secImproved
use ResidualPlanner instead of HDMM for reconstructionHDMM · Synth-nd, d=10reconstruction time: Out of memory in reconstruction -> 0.125 ± 0.032 secImproved
use ResidualPlanner instead of HDMM at very small dHDMM · Synth-nd, d=2reconstruction time: 0.003 ± 0.0006 sec -> 0.005 ± 0.003 secHurt
optimize max variance directly (ResidualPlanner) instead of RMSE-optimal HDMMHDMM · Adult, 3-way marginalsmax variance: 5675.238 -> 236.843Improved
optimize max variance directly (ResidualPlanner) instead of RMSE-optimal HDMMHDMM · CPS, <=3-way marginalsmax variance: 415.073 -> 13.216Improved
optimize max variance directly (ResidualPlanner) instead of RMSE-optimal HDMMHDMM · Loans, <=3-way marginalsmax variance: 4317.709 -> 180.817Improved
optimize max variance directly (ResidualPlanner) instead of RMSE-optimal HDMMHDMM · Adult, 1-way marginalsmax variance: 41.772 -> 12.047Improved

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