Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning to Combine: Knowledge Aggregation for Multi-Source Domain Adaptation
Hang Wang, Minghao Xu, Bingbing Ni, Wenjun Zhang
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 global relation alignment loss (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 86.90 -> 90.57 | Improved |
| add local relation alignment loss (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 86.90 -> 89.02 | Improved |
| add local loss on top of global relation alignment loss (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 90.57 -> 91.83 | Improved |
| add global loss on top of local relation alignment loss (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 89.02 -> 91.83 | Improved |
| add target entropy classification loss (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 87.17 -> 90.20 | Improved |
| add prototype classification loss (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 87.17 -> 88.43 | Improved |
| add prototype classification loss to source + target losses (ablation) | LtC-MSDA · Digits-five (average) | accuracy (%): 90.20 -> 91.83 | Improved |
| replace MDDA with LtC-MSDA | multi-source domain adaptation · Digits-five (->mm) | accuracy (%): 78.6 -> 85.6 | Improved |
| replace MDDA with LtC-MSDA | multi-source domain adaptation · DomainNet (average) | accuracy (%): 43.2 -> 47.4 | Improved |
| replace MDDA with LtC-MSDA | multi-source domain adaptation · Office-31 (average) | accuracy (%): 84.2 -> 84.6 | Improved |
| replace MDDA with LtC-MSDA | multi-source domain adaptation · PACS (average) | accuracy (%): 86.11 -> 89.85 | Improved |
| use LtC-MSDA vs MCD (source combine) on quickdraw task | multi-source domain adaptation · DomainNet (->qdr) | accuracy (%): 7.6 -> 16.3 | Improved |
Closest papers by meaning
- Cycle Self-Refinement for Multi-Source Domain Adaptation (AAAI 2024 · similarity 0.88)
- Collaborative Semantic Consistency Alignment for Blended-Target Domain Adaptation (AAAI 2025 · similarity 0.88)
- A Theory of Learning Unified Model via Knowledge Integration from Label Space Varying Domains (CVPR 2025 · similarity 0.87)
- Open-Set Graph Domain Adaptation via Separate Domain Alignment (AAAI 2024 · similarity 0.87)
- Invertible Projection and Conditional Alignment for Multi-Source Blended-Target Domain Adaptation (AAAI 2025 · similarity 0.86)
- Unveiling the Unknown: Unleashing the Power of Unknown to Known in Open-Set Source-Free Domain Adaptation (CVPR 2024 · similarity 0.86)
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.
