Research map / Label-efficient and robust learning / Domain adaptation and generalization
Cross-Domain Gradient Discrepancy Minimization for Unsupervised Domain Adaptation
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 |
|---|---|---|---|
| Add cross-domain gradient discrepancy minimization | CGDM, compared with MCD · DomainNet | mean accuracy: 6.7% | Helped |
| Add cross-domain gradient discrepancy minimization | CGDM, compared with SWD · DomainNet | mean accuracy: 3.6% | Helped |
| Add cross-domain gradient discrepancy minimization and self-supervised target learning | CGDM, compared with MCD · VisDA-2017 | mean accuracy: 10.4% | Helped |
| Add cross-domain gradient discrepancy minimization and self-supervised target learning | CGDM, compared with SWD · VisDA-2017 | mean accuracy: 5.9% | Helped |
| Add cross-domain gradient discrepancy minimization and self-supervised target learning | CGDM, compared with AFN · ImageCLEF | mean accuracy: 89.5% vs 88.9% | Helped |
| Add self-supervised learning for target samples (ablation) | Vanilla bi-classifier adversarial learning · ImageCLEF | mean accuracy | Helped |
| Add gradient discrepancy minimization (ablation) | Vanilla bi-classifier adversarial learning · ImageCLEF | mean accuracy | Helped |
| Combine self-supervised learning with gradient discrepancy minimization (ablation) | CGDM · ImageCLEF | mean accuracy | Helped |
| Use within-category conditional gradient alignment instead of marginal batch-level gradient alignment (ablation) | CGDM · UDA experiments | accuracy | No clear effect |
| Increase the gradient discrepancy loss weight β above 1 (ablation) | CGDM · ImageCLEF C→I | accuracy | Hurt |
| Set gradient discrepancy loss weight β between 0 and 1 (ablation) | CGDM · ImageCLEF C→I | accuracy | No clear effect |
Closest papers by meaning
- Can We Evaluate Domain Adaptation Models Without Target-Domain Labels? (ICLR 2024 · similarity 0.91)
- Re-Weighted Adversarial Adaptation Network for Unsupervised Domain Adaptation (CVPR 2018 · similarity 0.90)
- Bi-Classifier Determinacy Maximization for Unsupervised Domain Adaptation (AAAI 2021 · similarity 0.90)
- Bridging Theory and Algorithm for Domain Adaptation (ICML 2019 · similarity 0.90)
- Maximum Classifier Discrepancy for Unsupervised Domain Adaptation (CVPR 2018 · similarity 0.90)
- CASUAL: Conditional Support Alignment for Domain Adaptation with Label Shift (AAAI 2025 · 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.