Research map / Label-efficient and robust learning / Domain adaptation and generalization
Transferring Knowledge Fragments for Learning Distance Metric from a Heterogeneous Domain
Yong Luo, Yonggang Wen, Tongliang Liu, Dacheng Tao
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 |
|---|---|---|---|
| replace linear HTDML (RBF) with nonlinear GB-HTDML | HTDML · Scene-15 | accuracy: 0.452 -> 0.518 | Improved |
| add KPCA preprocessing to linear HTDML | HTDML · Scene-15 | accuracy: 0.452 -> 0.457 | Improved |
| add KPCA preprocessing to Euclidean baseline | EU · Scene-15 | accuracy: 0.409 -> 0.421 | Improved |
| add KPCA preprocessing to LMNN | LMNN · Scene-15 | accuracy: 0.435 -> 0.448 | Improved |
| add knowledge transfer from the source domain (ablation) | GB-HTDML · Scene-15 | accuracy | Improved |
| add manifold regularization (topology preservation) in the target domain (ablation) | GB-HTDML · Scene-15 | accuracy | Improved |
| increase number of unlabeled correspondences from 500 to 2000 (ablation) | GB-HTDML · Scene-15 | accuracy | Improved |
| increase number of unlabeled correspondences from 500 to 2000 (ablation) | DT and MI · Scene-15 | accuracy | No clear effect |
| use GB-HTDML instead of KPCA+MI | metric transfer · NUS-WIDE animal subset (20 labels per concept) | accuracy: 0.226 -> 0.270 | Improved |
| use GB-HTDML instead of KPCA+MI with more labels | metric transfer · NUS-WIDE animal subset (100 labels per concept) | accuracy: 0.270 -> 0.317 | Improved |
| use GB-HTDML instead of MI (labels plus correspondences) | metric transfer · LFW face verification | accuracy | Improved |
Closest papers by meaning
- Towards Improved Proxy-Based Deep Metric Learning via Data-Augmented Domain Adaptation (AAAI 2024 · similarity 0.83)
- A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning (NeurIPS 2025 · similarity 0.83)
- MergeNet: Knowledge Migration Across Heterogeneous Models, Tasks, and Modalities (AAAI 2025 · similarity 0.83)
- Knowledge Transfer from Vision Foundation Models for Efficient Training of Small Task-specific Models (ICML 2024 · similarity 0.83)
- A Unified Framework for Heterogeneous Semi-supervised Learning (CVPR 2025 · similarity 0.83)
- T-MDML: Triplet-based Multiple Distance Metric Learning for Multi-Instance Multi-Label Classification with Label Correlation (AAAI 2025 · similarity 0.83)
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.
