Research map / Label-efficient and robust learning / Domain adaptation and generalization
Representation Enhancement-Stabilization: Reducing Bias-Variance of Domain Generalization
Wei Huang, Yilei Shi, Zhitong Xiong, Xiao Xiang Zhu
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 feature-level frequency augmentation (amplitude noise, dropout, mixup) to ERM (ablation) | ERM · PACS | accuracy avg: 85.5 -> 88.8 | Improved |
| add MEMA mutual EMA stabilization to ERM (ablation) | ERM · PACS | accuracy std across runs: 1.1 -> 0.5 | Improved |
| combine RE and RS (ablation) | ERM · PACS | accuracy avg: 85.5 -> 90.0 (std 1.1 -> 0.3) | Improved |
| increase MEMA tau_S from 0.001 to 0.05 (ablation) | RES · TerraIncognita (test) | out-of-domain accuracy: 50.6 -> 40.2 | Hurt |
| increase MEMA tau_S from 0 to 0.05 (ablation) | RES · VLCS (test) | out-of-domain accuracy: 77.7 -> 79.8 | Improved |
| apply RES instead of ERM on PACS | ResNet-50 · PACS in-domain | accuracy avg: 96.5 -> 98.0 | Improved |
| RES versus best prior on VLCS and TerraInc | RES · VLCS, TerraIncognita | accuracy: 79.8 vs 80.0 (SAGM); 51.4 vs 51.5 (DomainDrop) | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- SWAD — SWAD: Domain Generalization by Seeking Flat Minima (NeurIPS 2021) · better on 3 of 3 shared results
- DomainDrop — DomainDrop: Suppressing Domain-Sensitive Channels for Domain Generalization (ICCV 2023) · better on 1 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
- Sharpness-Aware Model-Agnostic Long-Tailed Domain Generalization (AAAI 2024 · similarity 0.87)
- One-Step Generalization Ratio Guided Optimization for Domain Generalization (ICML 2025 · similarity 0.87)
- Towards Unsupervised Domain Generalization (CVPR 2022 · similarity 0.87)
- SelfReg: Self-Supervised Contrastive Regularization for Domain Generalization (ICCV 2021 · similarity 0.87)
- Failure Modes of Domain Generalization Algorithms (CVPR 2022 · similarity 0.87)
- Improving Generalization With Domain Convex Game (CVPR 2023 · 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.
