Research map / Label-efficient and robust learning / Domain adaptation and generalization
LAMDA: Label Matching Deep Domain Adaptation
Read the paperCheck 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 |
|---|---|---|---|
| Replace multi-class discriminator with binary discriminator in LAMDA (ablation) | LAMDA · Office-Home (ResNet-50) | Avg accuracy (%): 72.0 vs 68.7 (+3.3) | Helped |
| Reconstruction term weight beta (beta=0 used in real experiments) (ablation) | LAMDA · Office-31 (ResNet-50 features, ablation) | target test accuracy | No clear effect |
| Total transport cost term with weight alpha in [0.1, 0.5] (alpha=0.5 used) (ablation) | LAMDA · real-world datasets (ablation) | target test accuracy | Helped |
| Adding LAMDA (multi-class discriminator + optimal transport-inspired cost) over prior DDA methods | Office-Home · Office-Home ResNet-50 features | Avg accuracy (%): 72.0 vs 71.8 for SHOT (+0.2) and 46.1 for source-only | Helped |
| Adding LAMDA over prior methods | Office-31 · Office-31 ResNet-50 | Avg accuracy (%): 93.0 vs 90.8 for RWOT (+2.2) | Helped |
| Adding LAMDA over prior methods | digits, natural scenes · MNIST→SVHN | Accuracy (%): 98.4 vs 90.9 for SWD (+7.5); paper reports 10.7% margin over second best | Helped |
Closest papers by meaning
- RLSbench: Domain Adaptation Under Relaxed Label Shift (ICML 2023 · similarity 0.89)
- Label-Noise Robust Domain Adaptation (ICML 2020 · similarity 0.89)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.88)
- CASUAL: Conditional Support Alignment for Domain Adaptation with Label Shift (AAAI 2025 · similarity 0.87)
- Open-Set Heterogeneous Domain Adaptation: Theoretical Analysis and Algorithm (AAAI 2025 · similarity 0.87)
- COSDA: Counterfactual-based Susceptibility Risk Framework for Open-Set Domain Adaptation (ICML 2025 · similarity 0.87)
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.