Research map / Probabilistic and scientific modeling / Molecules and proteins
Measuring Dependence with Matrix-based Entropy Functional
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 |
|---|---|---|---|
| Use Iα* instead of the strongest listed pairwise dependence baseline | DREAM4 gene regulatory network inference · DREAM4 Networks 1–5 | AUC: Network 1: 0.75 -> 0.78; Network 2: 0.76 -> 0.87; Network 3: 0.83 -> 0.84; Network 5: 0.89 -> 0.97 | Helped |
| Train with Iα*(x;e) loss instead of HSIC loss | Fashion-MNIST classifier under covariate shift · Fashion-MNIST source and target | test accuracy (%): Source: 91.03±0.003 -> 91.17±0.040; Target: 76.56±0.034 -> 76.79±0.040 | Helped |
| Use Tα* or Dα* to select subspaces before LOF instead of full-space LOF | Subspace outlier detection · ODDS datasets | AUC: Diabetes: 0.68 vs 0.68; Cardio: 0.75 vs 0.68; Musk: 0.83 vs 0.42; Speech: 0.58 vs 0.36; breastW: 0.71 vs 0.46 | Mixed |
| Use DIB regularization instead of VIB | MNIST MLP · MNIST permutation-invariant | test error (%): 1.17 -> 1.13 | Helped |
| Use DIB regularization instead of VIB | VGG16 · CIFAR-10 | test error (%): 9.31 -> 5.66 | Helped |
| Replace HSIC loss with Hα(e) loss | Fashion-MNIST classifier under covariate shift · Fashion-MNIST source and target | test accuracy (%): Source: 91.03±0.003 -> 91.10±0.013; Target: 76.56±0.034 -> 75.48±0.069 | Mixed |
| Average Iα* estimates over random subsamples rather than using the full sample (ablation) | Iα* dependence test · Synthetic decaying-dependence test | dependence detection power | Hurt |
| Average Tα* estimates over random subsamples rather than using the full sample (ablation) | Tα* dependence score · Synthetic Data B | dependence score agreement with expected values | No clear effect |
Closest papers by meaning
- Sliced Mutual Information: A Scalable Measure of Statistical Dependence (NeurIPS 2021 · similarity 0.88)
- Robust and Fast Measure of Information via Low-Rank Representation (AAAI 2023 · similarity 0.88)
- On Slicing Optimality for Mutual Information (NeurIPS 2023 · similarity 0.88)
- Max-Sliced Mutual Information (NeurIPS 2023 · similarity 0.87)
- The Underlying Universal Statistical Structure of Natural Datasets (ICML 2025 · similarity 0.84)
- Connecting Jensen–Shannon and Kullback–Leibler Divergences: A New Bound for Representation Learning (NeurIPS 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.