Research map / Label-efficient and robust learning / Domain adaptation and generalization
Improved Knowledge Transfer for Semi-Supervised Domain Adaptation via Trico Training Strategy
Ba Hung Ngo, Yeon Jeong Chae, Jung Eun Kwon, Jae Hyeon Park, Sung In Cho
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 MCC loss to the MLP baseline (ablation) | baseline MLP classifier · DomainNet (3 tasks, 3-shot) | mean accuracy (%): 61.0 -> 67.5 | Improved |
| add co-training between MLP and inter-view GCN (ablation) | baseline + MCC · DomainNet (3 tasks, 3-shot) | mean accuracy (%): 61.0 -> 80.7 | Improved |
| add co-training between MLP and intra-view GCN (ablation) | baseline + MCC · DomainNet (3 tasks, 3-shot) | mean accuracy (%): 61.0 -> 77.4 | Improved |
| add third co-training between inter-view and intra-view GCN (ablation) | TriCT · DomainNet (3 tasks, 3-shot) | mean accuracy (%): 80.7 -> 83.6 | Improved |
| add MCC loss to full TriCT (ablation) | TriCT · DomainNet (3 tasks, 3-shot) | mean accuracy (%): 81.5 -> 83.6 | Improved |
| use TriCT instead of DEEM | DomainNet ResNet-34 · DomainNet (mean, 1-shot and 3-shot) | accuracy (%): 78.1 -> 80.7 (1-shot); 79.5 -> 85.5 (3-shot) | Improved |
| use TriCT instead of DEEM on DomainNet P->R 1-shot | DomainNet ResNet-34 · DomainNet P->R (1-shot) | accuracy (%): 85.4 -> 82.2 | Hurt |
| use TriCT instead of MCL on Office-Home 3-shot | Office-Home ResNet-34 · Office-Home (3-shot, mean) | accuracy (%): 77.1 -> 86.7 | Improved |
| use TriCT instead of CLDA on Office-31 AlexNet | Office-31 AlexNet · Office-31 (3-shot, mean) | accuracy (%): 71.5 -> 78.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CLDA — CLDA: Contrastive Learning for Semi-Supervised Domain Adaptation (NeurIPS 2021) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Attract, Perturb, and Explore: Learning a Feature Alignment Network for Semi-supervised Domain Adaptation (ECCV 2020 · similarity 0.88)
- Domain Adaptation With Auxiliary Target Domain-Oriented Classifier (CVPR 2021 · similarity 0.87)
- HiGDA: Hierarchical Graph of Nodes to Learn Local-to-Global Topology for Semi-Supervised Domain Adaptation (AAAI 2025 · similarity 0.87)
- Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias (CVPR 2024 · similarity 0.86)
- Enhancing Semi-supervised Domain Adaptation via Effective Target Labeling (AAAI 2024 · similarity 0.85)
- MLNet: Mutual Learning Network with Neighborhood Invariance for Universal Domain Adaptation (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.
