Research map / Learning theory and algorithms / Matrix and tensor methods
Multivariate Extension of Matrix-based Renyi's α-order Entropy Functional
Shujian Yu, Luis G. Sánchez Giraldo, Robert Jenssen, José Carlos Príncipe
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 |
|---|---|---|---|
| use full joint mutual information of the selected feature set instead of per-feature relevance only (MIM) | feature selection · 8 datasets | average rank: 6.13 -> 2.19 | Improved |
| use full joint mutual information of the feature subset rather than pairwise approximations (MRMR) | feature selection · 8 datasets | average rank: 3.50 -> 2.19 | Improved |
| lower alpha from 1.01 to 0.6 (ablation) | Ours · 8 datasets | average rank: 2.19 -> 1.63 | Improved |
| raise alpha from 1.01 to 2 (ablation) | Ours · 8 datasets | average rank: 2.19 -> 3.38 | Hurt |
| lower alpha to 0.6 (ablation) | Ours · Lymph | average rank: 3.54 -> 1.10 | Improved |
| lower alpha to 0.6 (ablation) | Ours · PIE | average rank: 1.90 -> 3.05 | Hurt |
| vary kernel size sigma over 0.5 to 10 (ablation) | Ours · waveform | average rank | No clear effect |
| vary kernel size sigma over 0.1 to 10 (ablation) | Ours · semeion | average rank | No clear effect |
| set sigma to 10 to 20 percent of median pairwise distance (ablation) | Ours · waveform and semeion | average rank | Improved |
| use already discretized data (Lung, Lymph) | Ours · Lymph | average rank: 3.54 | Hurt |
| use matrix-based joint MI with greedy selection | Ours vs JMI · Indian Pine 10% gallery | OA / AA | Hurt |
| use matrix-based joint MI with greedy selection | Ours vs others · Indian Pine 1% and 5% gallery | OA / AA | Improved |
Closest papers by meaning
- Nonlinear Laplacians: Tunable principal component analysis under directional prior information (NeurIPS 2025 · similarity 0.83)
- The Underlying Universal Statistical Structure of Natural Datasets (ICML 2025 · similarity 0.82)
- Spectral Estimation with Free Decompression (NeurIPS 2025 · similarity 0.82)
- Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability (NeurIPS 2025 · similarity 0.82)
- S$\Omega$I: Score-based O-INFORMATION Estimation (ICML 2024 · similarity 0.81)
- Structural-Entropy-Based Sample Selection for Efficient and Effective Learning (ICLR 2025 · 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.
