Research map / Label-efficient and robust learning / Domain adaptation and generalization
Min-Max Statistical Alignment for Transfer Learning
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 min-max alignment instead of minimizing disparity only (ablation) | UDA · Office31 (six transfer directions) | accuracy: Min-Max outperforms Min in all cases | Helped |
| use min-max alignment instead of minimizing disparity only (ablation) | UDA · MNIST→SVHN | accuracy: 49.8 -> 67.8 | Helped |
| use min-max alignment instead of minimizing disparity only (ablation) | UDA · remaining domain adaptation tasks | accuracy: Min-Max outperforms Min in all cases | Helped |
| add target-domain discriminative entropy loss to Min-Max (ablation) | UDA · non-Office31 domain adaptation tasks | accuracy | Helped |
| use min-max alignment instead of disparity minimization (ablation) | ZSL · Awa1 | HM: 59.3 -> 60.0 | Helped |
| use min-max alignment instead of disparity minimization (ablation) | ZSL · Awa2 | HM: 48.1 -> 53.0 | Helped |
| use min-max alignment instead of disparity minimization (ablation) | ZSL · CUB | HM: 48.3 -> 50.2 | Helped |
| use min-max alignment instead of disparity minimization (ablation) | ZSL · SUN | HM: 36.8 -> 37.2 | Helped |
| accumulate mini-batch means and covariances with momentum (ablation) | Min-Max alignment · Webcam→Amazon and CUB | performance | Helped |
| fine-tune the two-stage ZSL model end-to-end (ablation) | Min-Max ZSL · ZSL datasets | performance: < 1% | Helped |
| fine-tune the two-stage ZSL model end-to-end (ablation) | Min-Max ZSL · Awa1 | HM: 61.3% | Helped |
Closest papers by meaning
- Domain Adaptation with Invariant Representation Learning: What Transformations to Learn? (NeurIPS 2021 · similarity 0.89)
- Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation (ICLR 2018 · similarity 0.88)
- On Learning Domain-Invariant Representations for Transfer Learning with Multiple Sources (NeurIPS 2021 · similarity 0.88)
- Enhancing Domain Adaptation through Prompt Gradient Alignment (NeurIPS 2024 · similarity 0.88)
- Prompt-Based Distribution Alignment for Unsupervised Domain Adaptation (AAAI 2024 · similarity 0.87)
- Theoretical Performance Guarantees for Partial Domain Adaptation via Partial Optimal Transport (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.