Research map / Label-efficient and robust learning / Domain adaptation and generalization
From Big to Small: Adaptive Learning to Partial-Set Domains
Zhangjie Cao, Kaichao You, Ziyang Zhang, Jianmin Wang, Mingsheng Long
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 instance selection with shared layers to source-only ResNet (ablation) | ResNet · Office-31 PDA ablation (6 tasks avg) | accuracy (%): 87.05 -> 91.05 | Improved |
| add class selection (ablation) | SAN++ variant · Office-31 PDA ablation (6 tasks avg) | accuracy (%): 91.05 -> 93.24 | Improved |
| add entropy minimization (ablation) | SAN++ variant · Office-31 PDA ablation (6 tasks avg) | accuracy (%): 93.24 -> 95.32 | Improved |
| replace entropy minimization with self-training (ablation) | SAN++ · Office-31 PDA ablation (6 tasks avg) | accuracy (%): 95.32 -> 97.89 | Improved |
| share bottom layers of the multitask discriminator (ablation) | SAN++ · Office-31 PDA ablation (6 tasks avg) | accuracy (%): 97.02 -> 97.89 | Improved |
| share bottom layers of the multitask discriminator (ablation) | SAN++ · Office-31 I->C cost | peak GPU memory (MB): 5551.0 -> 569.5 | Improved |
| share bottom layers of the multitask discriminator (ablation) | SAN++ · Office-31 I->C cost | time (s): 1.601 -> 0.148 | Improved |
| add complement entropy (COT) loss | SAN++ · Office-Home PDA (12 tasks avg) | accuracy (%): 75.96 -> 77.37 | Improved |
| add complement entropy (COT) loss | SAN++ · VisDA-2017 PDA | accuracy (%): 63.06 -> 76.94 | Improved |
| add complement entropy (COT) loss | SAN++ · Office-31 PDA (6 tasks avg) | accuracy (%): 97.89 -> 98.18 | Improved |
| use SAN++ instead of SAN | SAN · Office-Home PDA (12 tasks avg) | accuracy (%): 65.30 -> 75.96 | Improved |
| use SAN++ instead of SAN | SAN · VisDA-2017 PDA | accuracy (%): 49.92 -> 63.06 | Improved |
Closest papers by meaning
- Partial Adversarial Domain Adaptation (ECCV 2018 · similarity 0.92)
- Learning to Transfer Examples for Partial Domain Adaptation (CVPR 2019 · similarity 0.90)
- Adaptive Adversarial Network for Source-Free Domain Adaptation (ICCV 2021 · similarity 0.89)
- A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation (ECCV 2020 · similarity 0.89)
- Rethinking the Role of Pre-Trained Networks in Source-Free Domain Adaptation (ICCV 2023 · similarity 0.89)
- Importance Weighted Adversarial Nets for Partial Domain Adaptation (CVPR 2018 · similarity 0.89)
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.
