Research map / Label-efficient and robust learning / Domain adaptation and generalization
Geometric Understanding of Discriminability and Transferability for Visual Domain Adaptation
You-Wei Luo, Chuan-Xian Ren, Xiaolin Xu, Qingshan Liu
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 geometry-oriented loss L_GO (ablation) | GOAL · Image-CLEF target representation | LDA discriminant value: 0.095 -> 1.194 | Improved |
| add geometry-oriented loss L_GO (ablation) | GOAL · Image-CLEF target representation | intra-class scatter: 0.912 -> 0.450 | Improved |
| add geometry-oriented loss L_GO (ablation) | GOAL · Image-CLEF target representation | principal angle cosine: 0.670 -> 0.994 | Improved |
| add geometry-oriented loss L_GO (ablation) | GOAL · Image-CLEF target representation | class-wise principal angle cosine: 0.707 -> 0.909 | Improved |
| add CDAN module to GOAL | GOAL · Office-31 | accuracy: 93.8 -> 94.1 | Improved |
| add CDAN module to GOAL | GOAL · Image-CLEF | accuracy: 91.3 -> 91.5 | Improved |
| add CDAN module to GOAL | GOAL · VisDA-2017 | accuracy: 83.7 -> 85.0 | Improved |
| add CDAN module to GOAL | GOAL · DomainNet | accuracy: 33.3 -> 34.9 | Improved |
| add CDAN module to GOAL | GOAL · Office-Home | accuracy: 68.6 -> 69.7 | Improved |
| set lambda_TB larger than lambda_DB (ablation) | GOAL · Office-Home, Office-31, Image-CLEF, VisDA-2017 | accuracy | Hurt |
| keep lambda within (0, 1+sqrt(2)] with balanced weights (ablation) | GOAL · Office-Home, Office-31, Image-CLEF, VisDA-2017 | accuracy | Improved |
| add warm-up training before geometric learning (ablation) | GOAL · Office-Home pseudo-labels | pseudo-label accuracy: about 10% ~ 14% gain | Improved |
Closest papers by meaning
- Joint Geometrical and Statistical Alignment for Visual Domain Adaptation (CVPR 2017 · similarity 0.88)
- Representation Subspace Distance for Domain Adaptation Regression (ICML 2021 · similarity 0.88)
- On the Transferability and Discriminability of Representation Learning in Unsupervised Domain Adaptation (TPAMI 2025 · similarity 0.88)
- Structure-Aware Feature Fusion for Unsupervised Domain Adaptation (AAAI 2020 · similarity 0.87)
- Cooperative and Adversarial Learning: Co-enhancing Discriminability and Transferability in Domain Adaptation (AAAI 2023 · similarity 0.87)
- Transferability vs. Discriminability: Batch Spectral Penalization for Adversarial Domain Adaptation (ICML 2019 · 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.
