Research map / Label-efficient and robust learning / Domain adaptation and generalization
S3VAADA: Submodular Subset Selection for Virtual Adversarial Active Domain Adaptation
Harsh Rangwani, Arihant Jain, Sumukh K Aithal, R. Venkatesh Babu
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 gradient clipping (norm 1) to VAADA training (ablation) | VAADA · Office-31 Webcam->Amazon | accuracy: above 10% per cycle | Improved |
| do not lower backbone learning rate (same rate as head) (ablation) | VAADA · Office-31 Webcam->Amazon | accuracy | Improved |
| use VAADA instead of VADA/DANN as adaptation procedure (ablation) | S3VAADA sampling · Office-31 Webcam->Amazon | accuracy: >= 5% per cycle | Improved |
| combine uncertainty, diversity and representativeness (alpha=0.5, beta=0.3) instead of uncertainty alone or diversity alone (ablation) | S3VAADA sampling · Office-31 Webcam->Amazon | accuracy | Improved |
| use S3VAADA sampling instead of other sampling methods under VAADA training | VAADA · Office-31 Webcam->Amazon | accuracy: 9% with 45 labeled samples | Improved |
| use S3VAADA instead of other methods on VisDA-18 | S3VAADA · VisDA-18 | accuracy: 7% averaged across cycles | Improved |
| use S3VAADA instead of other methods on easier Art->Product shift | S3VAADA · Office-Home Art->Product | accuracy | No clear effect |
| use S3VAADA instead of other methods on harder Office-Home shifts | S3VAADA · Office-Home Art->Clipart | accuracy: 3%-5% | Improved |
| increase number of active rounds beyond 15 | S3VAADA · Office-31 Webcam->Amazon | accuracy: 89% at 20 rounds | No clear effect |
| use very small labeling budget of 25 instead of >45 (ablation) | S3VAADA · Office-31 Webcam->Amazon | accuracy | Hurt |
| use S3VAADA instead of MME* semi-supervised DA with random labeling | S3VAADA · Office-31 | accuracy | Improved |
Closest papers by meaning
- Prototype-Driven Active Domain Adaptation with Density Consideration (AAAI 2026 · similarity 0.87)
- A DIRT-T Approach to Unsupervised Domain Adaptation (ICLR 2018 · similarity 0.87)
- A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation (ECCV 2020 · similarity 0.86)
- Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment (ICML 2019 · similarity 0.86)
- Model Adaptation: Unsupervised Domain Adaptation Without Source Data (CVPR 2020 · similarity 0.86)
- Category-Aware Active Domain Adaptation (ICML 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.
