Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Multi-Class Domain Adaptation: Theory, Algorithms, and Practice
Yabin Zhang, Bin Deng, Hui Tang, Lei Zhang, Kui Jia
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 |
|---|---|---|---|
| use CE surrogate instead of DANN-style binary domain discriminator (ablation) | McDalNets · Office-31 | accuracy (%): 82.8 -> 85.3 | Improved |
| use CE surrogate instead of DANN-style binary domain discriminator (ablation) | McDalNets · VisDA-2017 | accuracy (%): 58.4 -> 70.5 | Improved |
| use CE surrogate instead of MDD variant (ablation) | McDalNets · Office-Home | accuracy (%): 61.1 -> 64.0 | Improved |
| MDD variant (relative margin) on Digits and VisDA-2017 (ablation) | McDalNets · Digits | accuracy (%): not converge | Hurt |
| use CE surrogate instead of KL surrogate (ablation) | McDalNets · Digits | accuracy (%): 82.9 -> 94.9 | Improved |
| use CE surrogate instead of L1/MCD surrogate (ablation) | McDalNets · Office-Home | accuracy (%): 62.0 -> 64.0 | Improved |
| use CE surrogate instead of L1/MCD surrogate (ablation) | McDalNets · VisDA-2017 | accuracy (%): 70.4 -> 70.5 | No clear effect |
| use CE surrogate instead of L1/MCD surrogate (ablation) | McDalNets · DomainNet | accuracy (%): 27.7 -> 27.9 | No clear effect |
| replace best McDalNets surrogate (CE) with SymmNets-V2 domain confusion and discrimination (ablation) | SymmNets-V2 · Office-31 | accuracy (%): 85.3 -> 89.1 | Improved |
| replace best McDalNets surrogate (CE) with SymmNets-V2 (ablation) | SymmNets-V2 · Office-Home | accuracy (%): 64.0 -> 68.1 | Improved |
| replace best McDalNets surrogate (CE) with SymmNets-V2 (ablation) | SymmNets-V2 · DomainNet | accuracy (%): 27.9 -> 27.9 | No clear effect |
| replace best McDalNets surrogate (CE) with SymmNets-V2 (ablation) | SymmNets-V2 · VisDA-2017 | accuracy (%): 70.5 -> 71.3 | Improved |
Closest papers by meaning
- Bridging Theory and Algorithm for Domain Adaptation (ICML 2019 · similarity 0.91)
- Deep Cocktail Network: Multi-Source Unsupervised Domain Adaptation With Category Shift (CVPR 2018 · similarity 0.90)
- Distributionally Robust Classification for Multi-source Unsupervised Domain Adaptation (ICLR 2026 · similarity 0.90)
- Can We Evaluate Domain Adaptation Models Without Target-Domain Labels? (ICLR 2024 · similarity 0.90)
- Minimum Class Confusion for Versatile Domain Adaptation (ECCV 2020 · similarity 0.90)
- Domain-Symmetric Networks for Adversarial Domain Adaptation (CVPR 2019 · similarity 0.90)
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.
