Research map / Label-efficient and robust learning / Domain adaptation and generalization
Adjustment and Alignment for Unbiased Open Set Domain Adaptation
Wuyang Li, Jie Liu, Bo Han, Yixuan Yuan
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 FDA to debias source-domain learning (ablation) | ANNA · Office-Home | HOS: 65.9 -> 68.2 | Improved |
| Add full DCA with base- and novel-class alignment (ablation) | ANNA · Office-Home | HOS: 65.9 -> 69.0 | Improved |
| Add FDA together with both DCA alignment heads (ablation) | ANNA · Office-Home | HOS: 65.9 -> 70.3 | Improved |
| Add base-class alignment head (ablation) | ANNA DCA · Office-Home | OS*: 62.2 -> 63.7 | Improved |
| Add novel-class alignment head (ablation) | ANNA DCA · Office-Home | UNK: 70.9% -> 76.0% | Improved |
| Set FDA loss weight λ1 to 0.5 instead of 1.0 (ablation) | ANNA · Office-Home | HOS: 70.7 -> 69.5 | Hurt |
| Set FDA loss weight λ1 to 2.0 instead of 1.0 (ablation) | ANNA · Office-Home | HOS: 70.7% to 70.1% | Hurt |
| Set FDA loss weight λ1 to 2.0 instead of 1.0 (ablation) | ANNA · Image-CLEF | HOS: 81.4% to 81.2% | Hurt |
| Set DCA loss weight λ2 to 0.5 or 2.0 instead of 1.0 (ablation) | ANNA · Office-Home | HOS: 70.7 -> 69.3; 70.7 -> 69.6 | Hurt |
| Set DCA loss weight λ2 to 0.5 instead of 1.0 (ablation) | ANNA · Image-CLEF | HOS: 81.4 -> 79.0 | Hurt |
| Set DCA loss weight λ2 to 2.0 instead of 1.0 (ablation) | ANNA · Image-CLEF | HOS: 81.4 -> 81.3 | Hurt |
| Increase the novel-region selection threshold K too far (ablation) | ANNA FDA · Office-Home Ar→Cl | UNK | Hurt |
Closest papers by meaning
- Interpretable Open-Set Domain Adaptation via Angular Margin Separation (ECCV 2022 · similarity 0.90)
- Open-Set Heterogeneous Domain Adaptation: Theoretical Analysis and Algorithm (AAAI 2025 · similarity 0.90)
- COSDA: Counterfactual-based Susceptibility Risk Framework for Open-Set Domain Adaptation (ICML 2025 · similarity 0.90)
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.88)
- Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment (CVPR 2025 · similarity 0.87)
- Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias (CVPR 2024 · similarity 0.87)
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.
