Research map / Label-efficient and robust learning / Recognition and long-tail
Double-Weighting for Covariate Shift Adaptation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use double-weighting of training and test samples instead of KMM's training-only weighting | DW-GCS · 20 Newsgroups: five binary tasks | classification error: $.40\pm.02$ -> $.22\pm.02$; $.34\pm.03$ -> $.11\pm.02$; $.41\pm.02$ -> $.17\pm.02$; $.38\pm.03$ -> $.15\pm.02$; $.39\pm. | Helped |
| Use double-weighting of training and test samples rather than single-weight baselines | DW-GCS · Blood: Feature 1 and PCA | classification error: $.29\pm.04$ -> $.30\pm.03$; $.29\pm.05$ -> $.28\pm.04$ | Mixed |
| Use double-weighting of training and test samples rather than a single-weight baseline | DW-GCS · BreastCancer: Feature 1 and PCA | classification error: $.05\pm.02$ -> $.04\pm.02$; $.03\pm.01$ -> $.02\pm.01$ | Helped |
| Use double-weighting of training and test samples rather than RuLSIF | DW-GCS · Haberman: Feature 1 | classification error: $.31\pm.06$ -> $.28\pm.07$ | Helped |
| Use double-weighting of training and test samples rather than RuLSIF | DW-GCS · Ringnorm: PCA | classification error: $.25\pm.02$ -> $.26\pm.02$ | Hurt |
| Reduce the cap on training weights using D greater than 1 | DW-GCS · Covariate shift; theoretical effective sample size | effective sample size: D times larger | Helped |
Closest papers by meaning
- Covariate-Shift Generalization via Random Sample Weighting (AAAI 2023 · similarity 0.89)
- Sequential Covariate Shift Detection Using Classifier Two-Sample Tests (ICML 2022 · similarity 0.89)
- Positive-unlabeled AUC Maximization under Covariate Shift (ICML 2025 · similarity 0.85)
- Bayesian Weight Enhancement with Steady-State Adaptation for Test-time Adaptation in Dynamic Environments (ICML 2025 · similarity 0.85)
- Fairness under Covariate Shift: Improving Fairness-Accuracy Tradeoff with Few Unlabeled Test Samples (AAAI 2024 · similarity 0.85)
- High-Dimensional Kernel Methods under Covariate Shift: Data-Dependent Implicit Regularization (ICML 2024 · similarity 0.85)
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.