Research map / Label-efficient and robust learning / Domain adaptation and generalization
HoMM: Higher-Order Moment Matching for Unsupervised Domain Adaptation
Chao Chen, Zhihang Fu, Zhihong Chen, Sheng Jin, Zhaowei Cheng, Xinyu Jin, Xian-sheng Hua
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 |
|---|---|---|---|
| match third- and fourth-order feature moments, rather than only first- or second-order statistics | HoMM · Digits, Office-31, and Office-Home | accuracy | Improved |
| add pseudo-label-based discriminative clustering in target feature space (ablation) | HoMM · Digits, Office-31, and Office-Home | accuracy | Improved |
| use entropy regularization instead of pseudo-label-based clustering (ablation) | KHoMM · Digits and Office-31 | accuracy | Mixed |
| use kernelized HoMM instead of plain HoMM (ablation) | HoMM · Digits, Office-31, and Office-Home | accuracy | Improved |
| increase moment order from 3 to 4 (ablation) | HoMM · Office-31 and Office-Home | accuracy | Improved |
| increase moment order from 3 to 4 (ablation) | HoMM · Digits | accuracy | Hurt |
| increase moment order beyond 4 (ablation) | HoMM · SVHN→MNIST, Office-31 A→W, and Office-Home A→P | accuracy | Hurt |
Closest papers by meaning
- Moment Matching for Multi-Source Domain Adaptation (ICCV 2019 · similarity 0.89)
- Exploiting Local Feature Patterns for Unsupervised Domain Adaptation (AAAI 2019 · similarity 0.88)
- Unsupervised Domain Adaptation by Matching Distributions Based on the Maximum Mean Discrepancy via Unilateral Transformations (AAAI 2019 · similarity 0.88)
- Exact Gaussian Moment Matching for Residual Networks: a Second-Order Method (NeurIPS 2026 · similarity 0.85)
- Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs (NeurIPS 2026 · similarity 0.85)
- Tackling Dimensional Collapse toward Comprehensive Universal Domain Adaptation (ICML 2025 · similarity 0.85)
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.
