AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Distributed Algorithms for U-statistics-based Empirical Risk Minimization

Lanjue Chen, Alan T.K. Wan, Shuyi Zhang, Yong Zhou

JMLR 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization · 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
use SU-ERM/OS-ERM instead of naive averagingdistributed U-statistic ERM · Pairwise ranking, n=50, K 25 to 200RMSEImproved
use SU-ERM instead of naive averaging for confidence intervalsSU-ERM · Pairwise ranking theta*_1, K=200coverage probability: 0.115 -> 0.970Improved
use OS-ERM instead of naive averaging for confidence intervalsOS-ERM · Pairwise ranking theta*_1, K=25coverage probability: 0.755 -> 0.970Improved
use SU-ERM instead of OS-ERMSU-ERM · Pairwise rankingRMSENo clear effect
add iterations of DiaSER (ablation)DiaSER · Pairwise rankingRMSE: errors decrease rapidly within first two iterations then steadyImproved
increase number of machines K at fixed n=50naive estimator · Pairwise rankingRMSENo clear effect
increase K at fixed nSU-ERM and OS-ERM · Pairwise rankingcomputational timeNo clear effect
use SU-ERM instead of OS-ERMSU-ERM · Pairwise ranking and AFTcomputational timeHurt
use SU-ERM/OS-ERM instead of naive under AFT modeldistributed U-ERM · AFT model, n=50coverage probabilityImproved
decrease local sample size n at fixed N=6000 for naivenaive estimator · AFT model, N=6000coverage probability: naive CPs below 90% when n<60Hurt
decrease local sample size n at fixed NSU-ERM and OS-ERM · AFT model, N=6000, K=250 n=24coverage probability: fall below 90% only at (K,n)=(250,24)Hurt

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