Research map / Label-efficient and robust learning / Test-time adaptation
Sequential Covariate Shift Detection Using Classifier Two-Sample Tests
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 |
|---|---|---|---|
| Online training with per-sample accuracy evaluation (no held-out split) vs. held-out split baselines H2/H5 | covariate shift detector (classifier two-sample test) · ImageNet natural shift, M-shift | samples to reach >=80% detection: 190 vs 720 (H2) | Helped |
| Online detector vs. Deep Kernel baseline | covariate shift detector · ImageNet natural shift, M-shift | samples to reach >=80% detection: 190 (Ours) vs 180 (DK) | Mixed |
| Online detector vs. Deep Kernel baseline | covariate shift detector · ImageNet natural shift | FPR: DK FPR 2.54% vs Ours 0.77% at 200 samples | Hurt |
| Online detector with per-sample accuracy evaluation vs. held-out split baselines | covariate shift detector · ImageNet synthetic shift (Gaussian noise), M-shift and GI-shift | samples to reach >=80% detection: GI-shift 2080 (Ours) vs 3670 (H2) | Helped |
| Online detector vs. DK on synthetic shift M-shift | covariate shift detector · ImageNet synthetic shift (Gaussian noise), M-shift | samples to reach >=80% detection: 180 (Ours) vs 110 (DK) | Mixed |
| Online detector evaluated for false positive control | covariate shift detector · ImageNet synthetic shift, all scenarios | FPR: Ours FPR 0.75-0.99% within 1% bound | No clear effect |
Closest papers by meaning
- A Learning Based Hypothesis Test for Harmful Covariate Shift (ICLR 2023 · similarity 0.90)
- Double-Weighting for Covariate Shift Adaptation (ICML 2023 · similarity 0.89)
- Beyond Invariance: Test-Time Label-Shift Adaptation for Addressing "Spurious" Correlations (NeurIPS 2023 · similarity 0.87)
- Fairness under Covariate Shift: Improving Fairness-Accuracy Tradeoff with Few Unlabeled Test Samples (AAAI 2024 · similarity 0.86)
- Computational Efficiency under Covariate Shift in Kernel Ridge Regression (NeurIPS 2025 · similarity 0.84)
- Minimax Optimal Two-Stage Algorithm For Moment Estimation Under Covariate Shift (ICLR 2025 · similarity 0.84)
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.