Research map / Label-efficient and robust learning / Domain adaptation and generalization
Self-training Avoids Using Spurious Features Under Domain Shift
Yining Chen, Colin Wei, Ananya Kumar, Tengyu Ma
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add self-training on unlabeled target data | source classifier · CelebA | accuracy (%): 81\% -> 88\% | Improved |
| add self-training on unlabeled target data | source classifier · CMNIST10 (p = 0.95) | accuracy (%): 82\% -> 91\% | Improved |
| add self-training on unlabeled target data | source classifier · CMNIST2 | accuracy (%): 94\% -> 96\% | Improved |
| add self-training when source classifier is poor (spurious correlation raised to 0.97) | source classifier · CMNIST10 (p = 0.97) | accuracy (%): 72\% -> 67\% | Hurt |
| raise color-label correlation in source from 0.95 to 0.97 (ablation) | source classifier · CMNIST10 target | accuracy (%): 82\% -> 72\% | Hurt |
Closest papers by meaning
- Understanding Self-Training for Gradual Domain Adaptation (ICML 2020 · similarity 0.88)
- Semi-Supervised Domain Adaptation via Minimax Entropy (ICCV 2019 · similarity 0.88)
- Minimax optimal approaches to the label shift problem in non-parametric settings (JMLR 2022 · similarity 0.88)
- Label Noise: Ignorance Is Bliss (NeurIPS 2024 · similarity 0.87)
- Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs (NeurIPS 2026 · similarity 0.87)
- Out-Of-Domain Unlabeled Data Improves Generalization (ICLR 2024 · similarity 0.86)
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.
