Research map / Label-efficient and robust learning / Domain adaptation and generalization
Adversarial Style Mining for One-Shot Unsupervised Domain Adaptation
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 adversarial style mining instead of OST | one-shot domain adaptation · MNIST → SVHN | accuracy: 42.5 -> 46.3 | Helped |
| use adversarial style mining instead of OST | one-shot domain adaptation · USPS → SVHN | accuracy: 34.0 -> 40.3 | Helped |
| use adversarial style mining instead of OST | one-shot domain adaptation · MNIST → USPS | accuracy: 74.8 -> 68.0 | Hurt |
| use adversarial style mining instead of OST | one-shot segmentation · GTA5 → Cityscapes | mIoU: 42.3 -> 44.5 | Helped |
| use adversarial style mining instead of OST | one-shot segmentation · SYNTHIA → Cityscapes | mIoU: 42.8 -> 44.1 | Helped |
| add gradient-based adversarial style search rather than anchored sampling (ablation) | ASM · GTA5 → Cityscapes | mIoU: 42.9 -> 44.5 | Helped |
| add gradient-based adversarial style search rather than random sampling (ablation) | ASM · GTA5 → Cityscapes | mIoU: 42.4 -> 44.5 | Helped |
| replace style-focused adaptation with style transfer methods | classification domain adaptation · MNIST → USPS | accuracy: OST 74.8; ASM 68.0 | Mixed |
| reduce available target data from all samples to one while using ASM | ASM · GTA5 → Cityscapes | mIoU: 45.5 -> 44.5 | Hurt |
Closest papers by meaning
- Generalized One-shot Domain Adaptation of Generative Adversarial Networks (NeurIPS 2022 · similarity 0.90)
- Style Adaptation and Uncertainty Estimation for Multi-Source Blended-Target Domain Adaptation (NeurIPS 2024 · similarity 0.88)
- Adversarial Style Augmentation for Domain Generalized Urban-Scene Segmentation (NeurIPS 2022 · similarity 0.87)
- StyleAdv: Meta Style Adversarial Training for Cross-Domain Few-Shot Learning (CVPR 2023 · similarity 0.87)
- SVasP: Self-Versatility Adversarial Style Perturbation for Cross-Domain Few-Shot Learning (AAAI 2025 · similarity 0.86)
- On Unsupervised Domain Adaptation: Pseudo Label Guided Mixup for Adversarial Prompt Tuning (AAAI 2024 · similarity 0.85)
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.