Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Instance Specific Metric Subspace Learning: A Bayesian Approach
Han-Jia Ye, De-Chuan Zhan, Yuan Jiang
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 |
|---|---|---|---|
| ISMETSt vs Euclidean kNN | ISMETS · GDS3286 | test error: .441±.114 -> .338±.000 | Improved |
| ISMETSt vs SCMLG | ISMETS · balance | test error: .117±.032 -> .161±.021 | Hurt |
| ISMETSt vs SCMLG | ISMETS · echocardio | test error: .261±.059 -> .191±.031 | Improved |
| ISMETSt vs SCMLG | ISMETS · vote | test error: .142±.137 -> .066±.049 | Improved |
| ISMETSt vs ISD-L1 | ISMETS · GDS3286 | test error: .472±.145 -> .338±.000 | Improved |
| ISMETSt vs SDA | ISMETS · heart-stat | test error: .407±.090 -> .196±.038 | Improved |
| Transductive vs inductive prediction | ISMETS · GSE4115 | test error: .469±.019 -> .464±.026 | No clear effect |
| ISMETSt vs SDA | ISMETS · mfeat-mor | test error: .281±.015 -> .285±.018 | Hurt |
| Increase label ratio from 20% to 100% | Compared metric learning methods · haberman, heart-stat, mfeat-mor, echocardio | average error | Improved |
| Label ratio 20% (few labels) | ISMETSt/i · 4 datasets | average error | Improved |
| Vary number of metric bases K | ISMETSt/i · haberman, heart-stat | mean classification error | No clear effect |
| ISMETSi vs ISD-L1 and mmLMNN in training time | ISMETS · mfeat-mor | training speed: 7.82 and 1.72 times faster | Improved |
Closest papers by meaning
- Generative Local Metric Learning for Nearest Neighbor Classification (TPAMI 2017 · similarity 0.85)
- Variational Metric Scaling for Metric-Based Meta-Learning (AAAI 2020 · similarity 0.85)
- Learning Low-Dimensional Metrics (NeurIPS 2017 · similarity 0.84)
- Generative Local Metric Learning for Kernel Regression (NeurIPS 2017 · similarity 0.84)
- Efficient Stochastic Optimization for Low-Rank Distance Metric Learning (AAAI 2017 · similarity 0.83)
- Learning Distances from Data with Normalizing Flows and Score Matching (ICML 2025 · similarity 0.82)
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.
