Research map / Probabilistic and scientific modeling / Variational and generative sampling
Estimating Information Theoretic Measures via Multidimensional Gaussianization
Valero Laparra, J. Emmanuel Johnson, Gustau Camps‐Valls, Raúl Santos‐Rodríguez, Jesús Malo
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 RBIG instead of kNN for total correlation on Gaussian data | RBIG · Total correlation, Gaussian D=100 | relative MAE (%): 52.78 -> 1.55 | Improved |
| use RBIG instead of expF on rotated uniform data | RBIG · Total correlation, rotated uniform D=100 | relative MAE (%): 15.50 -> 7.50 | Improved |
| use RBIG instead of expF (Gaussian assumption) on Gaussian data | RBIG · Total correlation, Gaussian D=100 | relative MAE (%): 0.41 -> 1.55 | Hurt |
| use RBIG instead of kNN on Student nu=3 D=10 | RBIG · Total correlation, Student nu=3 D=10 | relative MAE (%): 16.73 -> 32.93 | Hurt |
| use RBIG instead of kNN for entropy on Gaussian D=100 | RBIG · Entropy, Gaussian D=100 | relative MAE (%): 30.70 -> 0.70 | Improved |
| use RBIG instead of kNN for KL divergence, Gaussian vs Student nu=2 D=100 | RBIG · KL divergence, Gaussian vs Student nu=2 D=100 | relative MAE (%): 1024.30 -> 42.11 | Improved |
| use RBIG instead of expF for KL divergence, Student vs Student nu=2 D=100 | RBIG · KL divergence, Student vs Student nu=2 D=100 | relative MAE (%): 35064.60 -> 242.80 | Improved |
| use RBIG instead of expF for KL divergence, Gaussians with different means D=100 | RBIG · KL divergence, Gaussians mu2=0.2 D=100 | relative MAE (%): 28.14 -> 47.40 | Hurt |
| use RBIG instead of kNN for mutual information, Gaussians D=100 | RBIG · Mutual information, Gaussians D=100 | relative MAE (%): 107.20 -> 11.70 | Improved |
| use RBIG instead of expF for mutual information, Student nu=20 D=100 | RBIG · Mutual information, Student nu=20 D=100 | relative MAE (%): 43.77 -> 37.21 | Improved |
Closest papers by meaning
- Estimators for Multivariate Information Measures in General Probability Spaces (NeurIPS 2018 · similarity 0.87)
- InfoNCE Induces Gaussian Distribution (ICLR 2026 · similarity 0.86)
- Partial Information Decomposition via Normalizing Flows in Latent Gaussian Distributions (NeurIPS 2025 · similarity 0.85)
- On the Importance of Gaussianizing Representations (ICML 2025 · similarity 0.84)
- Neural Mutual Information Estimation with Vector Copulas (NeurIPS 2025 · similarity 0.84)
- Analytically deriving Partial Information Decomposition for affine systems of stable and convolution-closed distributions (NeurIPS 2024 · 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.
