Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain-Shared Group-Sparse Dictionary Learning for Unsupervised Domain Adaptation
Baoyao Yang, Andy Ma, Pong Yuen
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 target-specific information (graph label propagation with target features) to the classifier (ablation) | DsGsDLNT · Office+Caltech (DeCAF6) average | accuracy (%): 90.29 -> 92.38 | Improved |
| add target-specific information (ablation) | DsGsDLNT · CMU-PIE source illumination, target blur sigma=3 | accuracy (%): 86.47 -> 88.82 | Improved |
| add target-specific information (ablation) | DsGsDLNT · CMU-PIE source illumination, target motion blur L=11 | accuracy (%): 77.74 -> 80.29 | Improved |
| add target-specific information (ablation) | DsGsDLNT · CMU-PIE source blur, target illumination | accuracy (%): 98.04 -> 98.82 | Improved |
| apply domain-shared group-sparse dictionary learning (without target-specific info) instead of source-only SVM (ablation) | SVM classifier · Office+Caltech (DeCAF6) average | accuracy (%): 84.73 -> 90.29 | Improved |
| estimate/align conditional distributions via label estimation or source selection (JDA, TJM) rather than none | domain adaptation on CMU-PIE · CMU-PIE face recognition | accuracy (%): TJM 59.71 vs SVMScr 73.82 (sigma=3) | Mixed |
| use domain-shared group-sparse alignment instead of JDA | domain adaptation · Office+Caltech (DeCAF6) average | accuracy (%): 85.45 -> 92.38 | Improved |
| use DsGsDL instead of deep DA methods | domain adaptation · Office+Caltech average | accuracy (%): 91.18 -> 92.38 | Improved |
Closest papers by meaning
- Unsupervised Domain Adaptation via Regularized Conditional Alignment (ICCV 2019 · similarity 0.87)
- Joint Domain Alignment and Discriminative Feature Learning for Unsupervised Deep Domain Adaptation (AAAI 2019 · similarity 0.86)
- Weakly Supervised Open-Set Domain Adaptation by Dual-Domain Collaboration (CVPR 2019 · similarity 0.86)
- Unsupervised Domain Adaptation Using Feature-Whitening and Consensus Loss (CVPR 2019 · similarity 0.86)
- Co-regularized Alignment for Unsupervised Domain Adaptation (NeurIPS 2018 · similarity 0.86)
- Link-based Contrastive Learning for One-Shot Unsupervised Domain Adaptation (CVPR 2025 · 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.
