Research map / Learning theory and algorithms / Kernels and regression theory
In Search of Lost Domain Generalization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 ERM with modern architectures, augmentation, and careful tuning instead of specialized domain-generalization algorithms | Domain-generalization algorithms · Seven DomainBed datasets | average accuracy | No clear effect |
| Use training-domain validation instead of leave-one-domain-out cross-validation for model selection | DomainBed algorithms · Multiple datasets | accuracy | Helped |
| Select models using test-domain validation (oracle) rather than training-domain validation | DomainBed algorithms · Seven DomainBed datasets | accuracy: +2% | Helped |
| Use larger ResNet-50 models, strong data augmentations, careful tuning, and the full Rotated MNIST training data | ERM · Domain generalization benchmarks | accuracy | Helped |
| Use ERM rather than the best prior reported method | Rotated MNIST · Rotated MNIST | average accuracy: 97.2 -> 98.0 | Helped |
| Use ERM rather than the best prior reported method | PACS · PACS | average accuracy: 84.5 -> 85.7 | Helped |
| Use ERM rather than the best prior reported method | VLCS · VLCS | average accuracy: 75.9 -> 77.4 | Helped |
| Use ERM rather than the best prior reported method | Office-Home · Office-Home | average accuracy: 65.5 -> 67.5 | Helped |
Closest papers by meaning
- Lost Domain Generalization Is a Natural Consequence of Lack of Training Domains (AAAI 2024 · similarity 0.90)
- Seeking Consistent Flat Minima for Better Domain Generalization via Refining Loss Landscapes (CVPR 2025 · similarity 0.90)
- Gradient-Guided Annealing for Domain Generalization (CVPR 2025 · similarity 0.90)
- Unknown Domain Inconsistency Minimization for Domain Generalization (ICLR 2024 · similarity 0.89)
- Set Valued Predictions For Robust Domain Generalization (ICML 2025 · similarity 0.89)
- Understanding Domain Generalization: A Noise Robustness Perspective (ICLR 2024 · 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.