AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Regression with Comparisons: Escaping the Curse of Dimensionality with Ordinal Information

Yichong Xu, Sivaraman Balakrishnan, Aarti Singh, Artur Dubrawski

JMLR 2020 · Label-efficient and robust learning › Semi-supervised learning · approach: Self-training theory · 5 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 pairwise comparison queries instead of induced comparisons from two direct label queriesAirBnB price estimation via AMT · AirBnB Task 1 (real price)worker error rate: 41.3% -> 31.3%Improved
use pairwise comparison queries instead of induced comparisons from two direct label queriesAirBnB price estimation via AMT · AirBnB Task 2 (user-estimated price)disagreement: 29.5% -> 16.4%Improved
ask a comparison query instead of a direct label queryAirBnB price estimation via AMT · AirBnB annotation timeaverage time per query: 63s -> 64sNo clear effect
add ranking (R^2 algorithm) over unlabeled samples to k-NN regressionk-NN regression · AirBnB price AMT, same total worker timeMSEImproved
add comparisons to linear regression (CLR) instead of labels onlylinear regression · AirBnB price AMTMSEImproved
use active instead of passive comparison queries in CLR (ablation)CLR · AirBnB Task 1MSENo clear effect
use active instead of passive comparison queries in CLR (ablation)CLR · AirBnB Task 2MSEImproved
add noisy comparisons instead of perfect comparisons to R^2 (cost ratio c >= 2) (ablation)R^2 on synthetic data · synthetic data, noisy comparisonsMSEMixed
increase number of comparisons beyond 50 in CLR (ablation)CLR · AirBnB (d=10)MSENo clear effect
use ranking when all samples are labeled (n=m) (ablation)R^2 vs k-NN · age estimation from imagesMSEImproved

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