Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Domain Adaptation by Matching Distributions Based on the Maximum Mean Discrepancy via Unilateral Transformations
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 |
|---|---|---|---|
| use residual transformation (A+I)x instead of unreferenced Ax (ablation) | MMD unilateral transformation · Office-Caltech10 and Amazon-Review (21 pairs) | accuracy (%): 42.42 -> 17.77 without residual on A->D; better on 21 of 21 | Helped |
| use polynomial kernel instead of RBF kernel (ablation) | MMD unilateral transformation · Office-Caltech10 W->D | accuracy (%): 76.82 -> 82.45 | Helped |
| use polynomial kernel instead of RBF kernel (ablation) | MMD unilateral transformation · Amazon-Review E->K | accuracy (%): 84.81 -> 84.75 | No clear effect |
| increase residual regularizer weight gamma from small values (ablation) | MMD unilateral transformation · Office-Caltech10 and Amazon-Review | mean accuracy | Helped |
| MMD-based unilateral transformation instead of no adaptation | linear SVM · Office-Caltech10 W->D | accuracy (%): 76.85 -> 82.45 | Helped |
| use CORAL alignment instead of no adaptation | linear SVM · Amazon-Review Ds->K | accuracy (%): 79.58 -> 76.39 | Hurt |
| use deep CMD trained from scratch instead of no adaptation | small CNN · Office-Caltech10 A->C | accuracy (%): 37.30 -> 28.86 | Hurt |
Closest papers by meaning
- Unsupervised Domain Adaptation With Distribution Matching Machines (AAAI 2018 · similarity 0.91)
- Learning Transferrable Representations for Unsupervised Domain Adaptation (NeurIPS 2016 · similarity 0.91)
- Co-regularized Alignment for Unsupervised Domain Adaptation (NeurIPS 2018 · similarity 0.91)
- Exploiting Local Feature Patterns for Unsupervised Domain Adaptation (AAAI 2019 · similarity 0.91)
- Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias (CVPR 2024 · similarity 0.89)
- Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs (NeurIPS 2026 · 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.