AI Paper Atlas

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

A Robust Learning Approach for Regression Models Based on Distributionally Robust Optimization

Ruidi Chen, Ioannis Ch. Paschalidis

JMLR 2018 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization

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 Wasserstein l2 (l2 regularizer) rather than l1 regularizer (ablation)Wasserstein DRO regression · Synthetic dense beta, outliers in x and yMSE / RR / RTE / PVEImproved
use Wasserstein l-infinity (l1 regularizer) rather than l2 when beta is sparse (ablation)Wasserstein DRO regression · Synthetic sparse beta, outliers in x and yMSE / RR / RTE / PVEImproved
penalize extended coefficient (-beta,1) instead of beta only (finite transport cost along y) (ablation)Wasserstein l2 DRO vs l2-regularized LAD · Synthetic dense beta, outliers only in xMSE / RR / RTE / PVEImproved
use absolute deviation loss instead of squared residual loss (ablation)Wasserstein DRO regression · Synthetic, large perturbations in ygeneralization and estimation errorImproved
use absolute deviation loss instead of squared residual loss when y noise is moderate (ablation)l2-regularized formulations · Synthetic dense beta, outliers only in xMSE / RR / RTE / PVEHurt
use Wasserstein DRO instead of l1-regularized LAD under varying regularization radiusoutlier detection · Outlier detection, out-of-sample AUC vs radiusAUCImproved
use Wasserstein DRO instead of Huber, Talwar, Fair M-estimation and l1-LADoutlier detection · Outlier cloud, N=60, sigma=0.5ROC curveImproved

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