Research map / Label-efficient and robust learning / Domain adaptation and generalization
Gradually Vanishing Bridge for Adversarial Domain Adaptation
Shuhao Cui, Shuhui Wang, Junbao Zhuo, Chi Su, Qingming Huang, Qi Tian
Check 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 GVB on the generator and discriminator | Adversarial domain adaptation · Office-31 (Avg) | accuracy (%): 87.0 -> {89.3} | Improved |
| Add GVB on the generator and discriminator | Adversarial domain adaptation · Office-Home (Avg) | accuracy (%): 68.3 -> 70.4 | Improved |
| Add GVB on the generator and discriminator | Adversarial domain adaptation · VisDA-2017 Synthetic → Real | accuracy (%): 71.3 -> 75.3 | Improved |
| Add GVB on the generator (ablation) | Office-Home ablation model · Office-Home (Avg; four tasks reported) | accuracy (%): 67.0 -> 68.5 | Improved |
| Add GVB on the discriminator instead of an unminimized discriminator bridge (ablation) | GVB-G on Office-Home · Office-Home (Avg; four tasks reported) | accuracy (%): 68.6 -> 69.1 | Improved |
| Use an unminimized discriminator bridge instead of GVB-D (ablation) | Discriminator bridge on Office-Home · Office-Home (Avg; four tasks reported) | accuracy (%): 67.8 -> 68.0 | No clear effect |
| Apply GVB to CDAN generator and discriminator | CDAN · Office-31 (Avg) | accuracy (%): 87.5 -> {88.8} | Improved |
| Apply GVB to CDAN generator and discriminator | CDAN · Office-Home (Avg) | accuracy (%): 65.8 -> 69.0 | Improved |
| Apply GVB to Symnets generator | Symnets · Office-31 (Avg) | accuracy (%): 88.4 -> 89.4 | Improved |
| Apply GVB to Symnets generator | Symnets · Office-Home (Avg) | accuracy (%): 67.2 -> 67.8 | Improved |
Closest papers by meaning
- Mind the Discriminability: Asymmetric Adversarial Domain Adaptation (ECCV 2020 · similarity 0.90)
- Adversarial Discriminative Domain Adaptation (CVPR 2017 · similarity 0.90)
- Generalized Domain Adaptation (CVPR 2021 · similarity 0.89)
- f-Domain Adversarial Learning: Theory and Algorithms (ICML 2021 · similarity 0.89)
- Reusing the Task-Specific Classifier as a Discriminator: Discriminator-Free Adversarial Domain Adaptation (CVPR 2022 · similarity 0.88)
- Domain Adversarial Training: A Game Perspective (ICLR 2022 · 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.
