AI Paper Atlas

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

The Robustness of Estimator Composition

Pingfan Tang, Jeff M Phillips

NeurIPS 2016 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization · 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
10% outliers in 10% of streams, E1 10% then E2 10% compositepercentile composite · Simulation 2command center output: -1.3327 -> -100.6573Improved
E1 10% / E2 90% composite needs ~10% outliers to be movedpercentile composite · Simulation 2command center output: -1.2169 -> 100.0553Improved
median-of-medians with 26.01% outliersmedian composite · Simulation 2command center output: -0.0085 -> 100.1062Improved
1.21% outliers on E1 10% / E2 median-type composite (median/median)median composite · Simulation 2command center output: -0.0085 -> 0.0091No clear effect
1.21% outliers on E1 10% / E2 90% compositepercentile composite · Simulation 2command center output: -1.2169 -> -1.2033No clear effect
raising outliers from 1.21% to 10.01% for E1 90%/E2 90%percentile composite · Simulation 2command center output: 100.5666 -> 108.6957Mixed
increase n and k from 5x8 to 1000x200 in L1-median modificationgradient descent algorithm · Simulation 1error in x0': 10.7631 -> 10.7025Mixed
n=1000, k=200 case vs smaller casesgradient descent algorithm · Simulation 1error in x0': 18.6351 -> 18.7886Hurt
tighter terminating condition (gradient norm < 0.00001)gradient descent algorithm · Simulation 1accuracy of target: 0.00001Improved
change only 12 of 40 pointsmodification algorithm · Simulation 1 n=5 k=8distance to target: 0.9961, 1.0126 -> 0.9961, 1.0126Improved

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