Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Domain Adaptation Using Feature-Whitening and Consensus Loss
Check 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 DWT-MEC instead of DWT alone (ablation) | DWT · CIFAR-10 → STL | accuracy (%): 79.75±0.25 -> 80.39±0.31 | Helped |
| Use DWT-MEC instead of DWT alone (ablation) | DWT · STL → CIFAR-10 | accuracy (%): 71.18±0.56 -> 72.52±0.94 | Helped |
| Add Mean-Teacher training to DWT-MEC (ablation) | DWT-MEC · CIFAR-10 → STL | accuracy (%): 80.39±0.31 -> 81.83±0.14 | Helped |
| Add Mean-Teacher training to DWT-MEC (ablation) | DWT-MEC · STL → CIFAR-10 | accuracy (%): 72.52±0.94 -> 71.31±0.22 | Hurt |
| Use MEC loss without confidence thresholding rather than SE consistency loss without thresholding (ablation) | DWT-MEC (MT) versus SE · SVHN → MNIST | accuracy (%): 26.80 -> 99.14 | Helped |
| Use MEC loss without confidence thresholding rather than SE consistency loss without thresholding (ablation) | DWT-MEC (MT) versus SE · MNIST → USPS | accuracy (%): 98.71 -> 99.30 | Helped |
| Use MEC loss without confidence thresholding rather than SE consistency loss without thresholding (ablation) | DWT-MEC (MT) versus SE · USPS → MNIST | accuracy (%): 97.63 -> 99.15 | Helped |
| Increase whitening group size from 1 toward 4 (ablation) | DWT · SVHN → MNIST | accuracy (%) | Helped |
| Increase whitening group size beyond 4 (ablation) | DWT · SVHN → MNIST | accuracy (%) | Hurt |
| Use more DWT layers (ablation) | DWT · SVHN → MNIST | accuracy (%) | Helped |
Closest papers by meaning
- Co-regularized Alignment for Unsupervised Domain Adaptation (NeurIPS 2018 · similarity 0.92)
- Unsupervised Domain Adaptation With Similarity Learning (CVPR 2018 · similarity 0.91)
- Exploiting Local Feature Patterns for Unsupervised Domain Adaptation (AAAI 2019 · similarity 0.91)
- Learning Transferrable Representations for Unsupervised Domain Adaptation (NeurIPS 2016 · similarity 0.91)
- Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation (ICLR 2018 · similarity 0.90)
- Enhancing Domain Adaptation through Prompt Gradient Alignment (NeurIPS 2024 · similarity 0.88)
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.