Research map / Label-efficient and robust learning / Domain adaptation and generalization
Transferable Attention for Domain Adaptation
Ximei Wang, Liang Li, Weirui Ye, Mingsheng Long, Jianmin Wang
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 transferable global attention to local-only (ablation) | TADA (local) · Office-31 Avg | accuracy: 84.5 -> 88.4 | Improved |
| add transferable local attention to global-only (ablation) | TADA (global) · Office-31 Avg | accuracy: 86.7 -> 88.4 | Improved |
| add transferable global attention to local-only (ablation) | TADA (local) · Office-Home Avg | accuracy: 63.9 -> 67.6 | Improved |
| add transferable local attention to global-only (ablation) | TADA (global) · Office-Home Avg | accuracy: 65.7 -> 67.6 | Improved |
| local attention only (no global) (ablation) | ResNet-50 adversarial DA · Office-31 D→A / W→A | accuracy: 72.9 -> 66.4 | Hurt |
| TADA vs MADA | adversarial domain adaptation · Office-31 Avg | accuracy: 85.2 -> 88.4 | Improved |
| TADA vs JAN | adversarial domain adaptation · Office-Home Avg | accuracy: 58.3 -> 67.6 | Improved |
| TADA vs SimNet | adversarial domain adaptation · Office-31 D→A | accuracy: 73.4 -> 72.9 | No clear effect |
Closest papers by meaning
- Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: the Benefit of Target Expectation Maximization (ECCV 2018 · similarity 0.90)
- Transferable Representation Learning with Deep Adaptation Networks (TPAMI 2018 · similarity 0.90)
- Transferable Adversarial Training: A General Approach to Adapting Deep Classifiers (ICML 2019 · similarity 0.89)
- Learning to Transfer Examples for Partial Domain Adaptation (CVPR 2019 · similarity 0.89)
- CrDoCo: Pixel-Level Domain Transfer With Cross-Domain Consistency (CVPR 2019 · similarity 0.88)
- Cooperative and Adversarial Learning: Co-enhancing Discriminability and Transferability in Domain Adaptation (AAAI 2023 · similarity 0.88)
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.
