Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning to Optimize Domain Specific Normalization for Domain Generalization
Seonguk Seo, Yumin Suh, Dongwan Kim, Geeho Kim, Jongwoo Han, Bohyung Han
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 |
|---|---|---|---|
| fine-tune BN parameters and statistics instead of fixing them (ablation) | ResNet-18 baseline · PACS avg | accuracy: 80.34 -> 78.86 | Hurt |
| add domain-specific optimized IN/BN mixture (DSON) | ResNet-18 baseline · PACS avg | accuracy: 78.86 -> 85.11 | Improved |
| add domain-specific optimized IN/BN mixture (DSON) | ResNet-50 baseline · PACS avg | accuracy: 82.48 -> 86.64 | Improved |
| add DSON | ResNet-18 baseline · Office-Home avg | accuracy: 61.96 -> 62.90 | Improved |
| add DSON | ResNet-18 baseline · Digits avg | accuracy: 85.28 -> 87.32 | Improved |
| use DSON instead of domain-specific BN (DSBN) (ablation) | ResNet-18 · PACS avg | accuracy: 77.61 -> 85.11 | Improved |
| use DSON instead of Switchable Normalization (SN) (ablation) | ResNet-18 · Office-Home avg | accuracy: 58.75 -> 62.90 | Improved |
| use DSON instead of IBN (ablation) | ResNet-18 · PACS avg | accuracy: 79.43 -> 85.11 | Improved |
| make IN/BN mixture weights domain-specific rather than shared (ablation) | DSON · PACS multi-source DG avg | accuracy: 82.82 -> 85.11 | Improved |
| use DSON under label noise 0.5 | ResNet-18 · PACS label noise 0.5 avg | accuracy: 71.59 -> 81.59 | Improved |
| add DSON to multi-source domain adaptation baseline | Shu et al. baseline · PACS multi-source DA avg | accuracy: 83.30 -> 90.38 | Improved |
| use DSON instead of DSBN in multi-source domain adaptation (ablation) | Shu et al. baseline · PACS multi-source DA avg | accuracy: 87.32 -> 90.38 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MetaReg — MetaReg: Towards Domain Generalization using Meta-Regularization (NeurIPS 2018) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Adversarially Adaptive Normalization for Single Domain Generalization (CVPR 2021 · similarity 0.92)
- Domain-Specific Batch Normalization for Unsupervised Domain Adaptation (CVPR 2019 · similarity 0.91)
- Customizing Domain Adapters for Domain Generalization (ICCV 2025 · similarity 0.90)
- Efficient Domain Generalization via Common-Specific Low-Rank Decomposition (ICML 2020 · similarity 0.89)
- Model-Based Domain Generalization (NeurIPS 2021 · similarity 0.89)
- Reducing Divergence in Batch Normalization for Domain Adaptation (AAAI 2025 · 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.
