Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Generalization Using a Mixture of Multiple Latent Domains
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 the proposed latent-domain method instead of JiGen | AlexNet · PACS | accuracy (Avg.): 73.38 -> 74.38 | Helped |
| use the proposed latent-domain method instead of D-SAM | ResNet-18 · PACS | accuracy (Avg.): 80.72 -> 81.83 | Helped |
| use the proposed latent-domain method instead of JiGen | AlexNet · VLCS | accuracy (Avg.): 73.19 -> 73.88 | Helped |
| add adversarial loss (ablation) | Ours · PACS | accuracy (Avg.): 72.40 -> 74.38 | Helped |
| add entropy loss (ablation) | Ours · PACS | accuracy (Avg.): 73.52 -> 74.38 | Helped |
| use stacked convolutional feature statistics for clustering rather than convolutional outputs (ablation) | Ours · PACS | accuracy (Avg.): 72.48 -> 74.38 | Helped |
| iteratively reassign pseudo domain labels during training (ablation) | Ours · PACS | accuracy (Avg.): 73.59 -> 74.38 | Helped |
| cluster style features for pseudo domain labels rather than use original domain labels (ablation) | Ours · PACS | accuracy (Avg.): 74.08 -> 74.38 | Helped |
| vary the number of pseudo domains (ablation) | Ours · PACS | classification accuracy | No clear effect |
Closest papers by meaning
- Domain Generalization With Adversarial Feature Learning (CVPR 2018 · similarity 0.90)
- Boosting Domain Adaptation by Discovering Latent Domains (CVPR 2018 · similarity 0.90)
- Domain Invariant Representation Learning with Domain Density Transformations (NeurIPS 2021 · similarity 0.90)
- Domain Generalization via Entropy Regularization (NeurIPS 2020 · similarity 0.89)
- Exploiting Domain-Specific Features to Enhance Domain Generalization (NeurIPS 2021 · similarity 0.89)
- Adversarial Domain Prompt Tuning and Generation for Single Domain Generalization (CVPR 2025 · 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.