Research map / Learning theory and algorithms / Kernels and regression theory
Trace Ratio Optimization With Feature Correlation Mining for Multiclass Discriminant Analysis
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 |
|---|---|---|---|
| tune alpha (l2,1 regularisation) instead of alpha=0 (no regularisation) (ablation) | TRLN · orlraws10P | 1NN accuracy (%): almost 10% | Helped |
| set alpha=0.001 rather than best at dim 3 (ablation) | TRLN · orlraws10P m=3 | 1NN accuracy (%): 86.40 vs 83.77 | Mixed |
| add l2,1 regularisation at pixraw10P m=1 (ablation) | TRLN · pixraw10P m=1 | 1NN accuracy (%): 53.80 | No clear effect |
| vary reduced dimension m from 9 down to 1 (ablation) | TRLN · orlraws10P | 1NN accuracy (%): 98.60 -> 42.60 | Hurt |
| vary reduced dimension m from 9 down to 1 (ablation) | TRLN · pixraw10P | 1NN accuracy (%): 99.80 -> 53.80 | Hurt |
| TRLN vs FLDA, aPAC, HMSS, GMSS at m=c-1 | TRLN · CLL SUB 111 | 1NN accuracy (%): almost 11% | Helped |
| TRLN vs other methods at m=c-1 | TRLN · TOX 171 | 1NN accuracy (%): almost 14% | Helped |
| TRLN vs other methods at m=c-1, linear SVM | TRLN · CLL SUB 111 | 1NN accuracy (%): 9% | Helped |
| TRLN vs other methods at m=c-1, linear SVM | TRLN · TOX 171 | 1NN accuracy (%): 15% | Helped |
| TRLN vs other methods at best dimension | TRLN · Carcinom | 1NN accuracy (%): almost 5% | Helped |
| TRLN vs other methods at best dimension | TRLN · Yale | 1NN accuracy (%): almost 5% | Helped |
| TRLN vs other methods at best dimension | TRLN · GLIOMA | 1NN accuracy (%): almost 4% | Helped |
Closest papers by meaning
- Ratio Trace Formulation of Wasserstein Discriminant Analysis (NeurIPS 2020 · similarity 0.88)
- Neural Fisher Discriminant Analysis: Optimal Neural Network Embeddings in Polynomial Time (ICML 2022 · similarity 0.85)
- Complex Moment-Based Supervised Eigenmap for Dimensionality Reduction (AAAI 2019 · similarity 0.85)
- Latent Discriminant Analysis with Representative Feature Discovery (AAAI 2017 · similarity 0.84)
- Heterogeneous Sufficient Dimension Reduction and Subspace Clustering (ICML 2025 · similarity 0.81)
- Rank-Transformed Dissimilarity Profiles for High-Dimensional Classification (NeurIPS 2026 · similarity 0.81)
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.