AI Paper Atlas

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

Doubly Debiased Robust Subsampling for Transfer Learning

Tao Wang, Weng Kee Wong

JMLR 2026 · 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
add target refinement to IPW subsampling (ablation)Doubly debiased robust subsampling · DGP1 covariate shift, Delta=0.4, nS=50K nT=2Ktest MSE: 1.2107 -> 1.1429Improved
add IPW debiasing to target refinement (ablation)Doubly debiased robust subsampling · DGP1 covariate shift, Delta=0.4, nS=50K nT=2Ktest MSE: 1.1764 -> 1.1429Improved
add target refinement to IPW subsampling (ablation)Doubly debiased robust subsampling · DGP2 label shift, severe, nS=50K nT=2Kmisclassification error (%): 25.1862 -> 22.8763Improved
add IPW debiasing to target refinement (ablation)Doubly debiased robust subsampling · DGP2 label shift, severe, nS=50K nT=2Kmisclassification error (%): 24.6237 -> 22.8763Improved
add target refinement to IPW subsampling (ablation)Doubly debiased robust subsampling · DGP3 concept shift, zeta=0.2, nS=50K nT=2Ktest MSE: 1.3296 -> 1.1964Improved
add IPW debiasing to target refinement (ablation)Doubly debiased robust subsampling · DGP3 concept shift, zeta=0.2, nS=50K nT=2Ktest MSE: 1.2586 -> 1.1964Improved
add target refinement to IPW subsampling (ablation)Doubly debiased robust subsampling · DGP4 heavy tails, nu=2, nS=50K nT=2Ktest MSE: 2.0326 -> 1.7826Improved
use the proposed method instead of uniform subsamplingDoubly debiased robust subsampling · DGP1 covariate shift, Delta=0.4, nS=50K nT=2Ktest MSE: 1.2897 -> 1.1429Improved
use the proposed method instead of target-only trainingDoubly debiased robust subsampling · DGP1 covariate shift, Delta=0.0, nS=50K nT=2Ktest MSE: 1.0026 -> 0.9918Improved
use the proposed method instead of IWERMDoubly debiased robust subsampling · DGP2 label shift, severe, nS=50K nT=2Kmisclassification error (%): 24.9731 -> 22.8763Improved
use leverage-score subsampling instead of uniformDoubly debiased robust subsampling · DGP4 heavy tails, nu=2, nS=50K nT=2Ktest MSE: 2.1047 -> 2.1869Hurt
increase target sample size nT from 2K to 5KDoubly debiased robust subsampling · DGP1 covariate shift, Delta=0.4, nS=50Ktest MSE (proposed): 1.1429 -> 1.1017Improved

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