โ€บNavigation
Tier A ยท Measurement ยท A1
PHASE 127 ENTRIESTIER A ยท MEASUREMENT

๐Ÿ”— A1 โ€” Pairwise dependence (incl. MI estimators)

Top pick: Chatterjee ฮพโ‚™ (xicorpy (MIT) โ€” VALIDATED (Selection-partial)). Fills the gap: Non-linear redundancy (ฮพ: S1/S2/S3/S5 PASS; value-add pending S4) โ€” which the production baseline (Pearson + Spearman + h_norm + Tier-2 MI-vs-target) cannot detect.

Total
27
metrics in this category
Permissive FOSS
11
usable today
GPL
10
license check needed
Unpackaged / no FOSS
3
integration cost
FOSS-READINESS LANES click any metric circle to open its library/repo URL

Catalog entries grouped by adoption-readiness

Lane = how usable the metric is today. Top lane = production-baseline duplicates (Pearson, Spearman). Below = permissive FOSS / GPL / unmaintained / code-released-but-not-packaged / no-FOSS. Per-circle text = first 3 letters of language (Pyt / R / Jav).

PRODUCTION BASELINE
FOSS โ€” permissive
FOSS โ€” GPL
FOSS โ€” unmaintained
Code released, not packaged
No FOSS surfaced

Full metric inventory

27 entries ยท sorted by FOSS readiness then name
Metric Library License Lang Maint Complexity Param-free? URL
Robust dCor (Leyder 2024)Author code (not packaged)variesRโš ๏ธO(N^2)noโ†—
Bergsma-Dassios tau-starTauStarGPLRโœ…O(N log N)noโ†—
Chatterjee xi โ€” RXICORGPLRโœ…O(N log N)noโ†—
Distance Correlation (dCor) โ€” RenergyGPLRโœ…O(N^2)noโ†—
Distance MultivariancemultivarianceGPLRโœ…O(N^2) per cdmnoโ†—
HHGHHGGPLRโœ…O(N^2 log N) per permyesโ†—
IDTxlIDTxlGPLv3Pythonโœ…variesyesโ†—
JIDT (KSG + many estimators)jidtGPLv3Java+Pythonโœ…variesyesโ†—
MIC / MICe / TICminepy/mictoolsGPLPython+Cโœ…polynomial; MICe efficientyesโ†—
Schweizer-Wolff sigmacopBasicGPLRโœ…O(N^2)noโ†—
dHSICdHSICGPLRโœ…O(N^2 d)yesโ†—
Chatterjee xixicorpyMITPythonโœ…O(N log N)noโ†—
Copula EntropycopentMITPythonโœ…O(N log N)noโ†—
Distance Correlation (dCor)dcorMITPythonโœ…O(N^2) naive; O(N log N) univariate Huo-Szekelynoโ†—
HSIChyppo.independence.HsicApache-2Pythonโœ…O(N^2); Nystrom sub-quadraticyesโ†—
HSIC-LassopyHSICLassoMITPythonโœ…O(d N^2) memory vanillayesโ†—
Hoeffding DindependenceMITRโœ…O(N log N)noโ†—
KSG MI estimatorinfomeasureMITPythonโœ…O(N log N)yesโ†—
Kendall tauscipy.stats.kendalltauBSD-3Pythonโœ…O(N log N)noโ†—
MINE (neural MI)MINEMITPythonโœ…O(N epochs params)yesโ†—
Sliced Wasserstein primitivesPOTMITPythonโœ…O(N log N d L)yesโ†—
Tail dependence (lambda_U, lambda_L)pycopMITPythonโœ…O(N)yesโ†—
CLUB (MI upper bound)CLUBMITPythonโš ๏ธO(N params)yesโ†—
DiPMIndnot releasedn/an/aโš ๏ธO(N^2)yesโ†—
Lin-Han boosted xiOn-boosting-the-power-of-Chatterjee-s-rank-correlationunknownPythonโš ๏ธO(N log N)yesโ†—
Pearson rscipy.stats.pearsonrBSD-3Pythonโœ…O(N)noโ†—
Spearman rhoscipy.stats.spearmanrBSD-3Pythonโœ…O(N log N)noโ†—

Cross-references