Research map / Label-efficient and robust learning / Domain adaptation and generalization
A Causal Inspired Early-Branching Structure for Domain Generalization
Liang Chen, Yong Zhang, Yibing Song, Zhen Zhang, Lingqiao Liu
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 |
|---|---|---|---|
| branch early (share only first block) instead of deep shared extractor (ablation) | dual-branch HSIC framework · PACS (ResNet18) | avg accuracy: 78.3 -> 81.0 | Improved |
| branch early instead of deep shared extractor (ablation) | dual-branch HSIC framework · PACS (ResNet50) | avg accuracy: 81.7 -> 84.7 | Improved |
| branch early instead of deep shared extractor (ablation) | dual-branch HSIC framework · RMNIST (4-conv network) | avg accuracy: 95.9 -> 98.0 | Improved |
| branch early instead of deep shared extractor (ablation) | dual-branch HSIC framework · PACS (DeiT) | avg accuracy: 83.6 -> 86.4 | Improved |
| add RDS augmentation with consistency (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.0 -> 82.4 | Improved |
| use RDS instead of MixStyle (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.9 -> 82.4 | Improved |
| use RDS instead of DSU (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.7 -> 82.4 | Improved |
| add low-density acceptance rule in RDS sampling (vs naive sampling RDS-) (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.9 -> 82.4 | Improved |
| add HSIC independence loss (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.8 -> 82.4 | Improved |
| add consistency loss (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.9 -> 82.4 | Improved |
| add MLP in consistency loss (ablation) | Ours · PACS (ResNet18) | avg accuracy: 81.9 -> 82.4 | Improved |
| use HSIC instead of correlation constraint (ablation) | Ours · PACS (ResNet18) | avg accuracy: 82.2 -> 82.4 | Improved |
Closest papers by meaning
- Causality Inspired Representation Learning for Domain Generalization (CVPR 2022 · similarity 0.91)
- Integrating Markov Blanket Discovery into Causal Representation Learning for Domain Generalization (ECCV 2024 · similarity 0.89)
- Enhancing Evolving Domain Generalization through Dynamic Latent Representations (AAAI 2024 · similarity 0.88)
- Minimal Semantic Sufficiency Meets Unsupervised Domain Generalization (NeurIPS 2025 · similarity 0.87)
- Robust Domain Generalization under Divergent Marginal and Conditional Distributions (NeurIPS 2026 · similarity 0.87)
- Towards Generalizing to Unseen Domains with Few Labels (CVPR 2024 · similarity 0.87)
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.
