Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Generalization With Adversarial Feature Learning
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 |
|---|---|---|---|
| combine MMD alignment and adversarial prior matching with supervised training | MMD-AAE · VLCS object recognition (average) | accuracy: 70.2 -> 72.3 | Helped |
| add adversarial prior matching to MMD-AAE (ablation) | MMD-AAE · VLCS object recognition | accuracy: 66.1 -> 67.7 | Helped |
| add MMD domain alignment (ablation) | MMD-AAE · VLCS object recognition | accuracy: 65.9 -> 67.7 | Helped |
| incorporate labels with the joint classification layer (ablation) | MMD-AAE · VLCS object recognition | accuracy: 67.1 -> 67.7 | Helped |
| replace the Laplace prior with a Gaussian prior N(0, 0.1 × I) (ablation) | MMD-AAE · VLCS object recognition | accuracy: 65.9 -> 67.7 | Hurt |
| replace the Laplace prior with a Uniform prior [−1, 1] (ablation) | MMD-AAE · VLCS object recognition | accuracy: 65.4 -> 67.7 | Hurt |
| use a narrower Laplace prior bandwidth of 10⁻² instead of 1/√2 (ablation) | MMD-AAE · VLCS object recognition | accuracy: 67.7 -> 68.2 | Helped |
| use a narrower Laplace prior bandwidth of 10⁻¹ instead of 1/√2 (ablation) | MMD-AAE · VLCS object recognition | accuracy: 67.7 -> 68.8 | Helped |
| use an extremely narrow Laplace prior bandwidth of 10⁻⁴ instead of 1/√2 (ablation) | MMD-AAE · VLCS object recognition | accuracy: 67.7 -> 65.8 | Hurt |
| use a narrow Gaussian prior N(0, 10⁻³ × I) instead of the Laplace prior (ablation) | MMD-AAE · VLCS object recognition | accuracy: 67.7 -> 67.8 | Helped |
Closest papers by meaning
- Learning to Learn Single Domain Generalization (CVPR 2020 · similarity 0.92)
- Domain Invariant Representation Learning with Domain Density Transformations (NeurIPS 2021 · similarity 0.91)
- Generalizing to Unseen Domains via Adversarial Data Augmentation (NeurIPS 2018 · similarity 0.91)
- Domain Generalization via Entropy Regularization (NeurIPS 2020 · similarity 0.91)
- Domain Generalization Using a Mixture of Multiple Latent Domains (AAAI 2020 · similarity 0.90)
- Sharpness-Aware Model-Agnostic Long-Tailed Domain Generalization (AAAI 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.