Research map / Label-efficient and robust learning / Domain adaptation and generalization
Blending-Target Domain Adaptation by Adversarial Meta-Adaptation Networks
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 AMEAN instead of the strongest listed baseline | Digit-five, Backbone-1 · Digit-five, average | ACC (%): 58.5 -> 61.5 | Helped |
| Use AMEAN instead of the strongest listed baseline | Digit-five, Backbone-2 · Digit-five, average | ACC (%): 72.3 -> 80.0 | Helped |
| Use AMEAN instead of the strongest listed baseline | Office-31, AlexNet · Office-31, average | ACC (%): 64.7 -> 65.7 | Helped |
| Use AMEAN instead of the strongest listed baseline | Office-31, ResNet-50 · Office-31, average | ACC (%): 76.9 -> 80.2 | Helped |
| Use AMEAN instead of the strongest listed baseline | Office-Home, AlexNet · Office-Home, average | ACC (%): 45.0 -> 46.3 | Helped |
| Use AMEAN instead of the strongest listed baseline | Office-Home, ResNet-50 · Office-Home, average | ACC (%): 57.9 -> 64.0 | Helped |
| Replace explicit sub-target guidance with the learned meta-learner (ablation) | AMEAN, Digit-five, mt→mm,sv,up,sy · Digit-five, mt→mm,sv,up,sy | ACC (%): 79.9 -> 85.1 | Helped |
| Replace static initial k-clusters with dynamically updated meta-sub-targets (ablation) | AMEAN, Digit-five, mt→mm,sv,up,sy · Digit-five, mt→mm,sv,up,sy | ACC (%): 81.6 -> 85.1 | Helped |
| Replace explicit sub-target guidance with the learned meta-learner (ablation) | AMEAN, Digit-five, mm→mt,sv,up,sy · Digit-five, mm→mt,sv,up,sy | ACC (%): 74.7 -> 77.6 | Helped |
| Replace static initial k-clusters with dynamically updated meta-sub-targets (ablation) | AMEAN, Digit-five, mm→mt,sv,up,sy · Digit-five, mm→mt,sv,up,sy | ACC (%): 74.5 -> 77.6 | Helped |
| Replace explicit sub-target guidance with the learned meta-learner (ablation) | AMEAN, Office-31, D→A,W · Office-31, D→A,W | ACC (%): 62.2 -> 62.8 | Helped |
| Replace static initial k-clusters with dynamically updated meta-sub-targets (ablation) | AMEAN, Office-31, D→A,W · Office-31, D→A,W | ACC (%): 61.3 -> 62.8 | Helped |
Closest papers by meaning
- Style Adaptation and Uncertainty Estimation for Multi-Source Blended-Target Domain Adaptation (NeurIPS 2024 · similarity 0.91)
- Adversarial-Learned Loss for Domain Adaptation (AAAI 2020 · similarity 0.90)
- Transferable Adversarial Training: A General Approach to Adapting Deep Classifiers (ICML 2019 · similarity 0.90)
- Learning to Transfer Examples for Partial Domain Adaptation (CVPR 2019 · similarity 0.90)
- Invertible Projection and Conditional Alignment for Multi-Source Blended-Target Domain Adaptation (AAAI 2025 · similarity 0.89)
- Can We Evaluate Domain Adaptation Models Without Target-Domain Labels? (ICLR 2024 · 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.