Mutual Information Estimators

Provides mutual information estimators based on k-nearest neighbor estimators by A. Kraskov, et al. (2004) , S. Gao, et al. (2015) <> and local density estimators by W. Gao, et al. (2017) .

An R package for kNN mutual information estimators. Mutual information is an information theoretic quantity that measures the amount of information that knowledge of one random variable provides about another random variable. The estimators contained in this package estimate the empirical mutual information between two or more variable from samples. Mutual information is useful for measuring nonlinear dependency between variables and applied to machine learning, signal processing, independent component analysis, and experimental design.


You can install rmi from github with:



rmi 0.1.1

  • Initial public release of rmi package
  • Added KSG1, KSG2, LNN, and LNC estimators
  • Added method to tune LNC parameters
  • Added a file to track changes to the package.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.1.1 by Isaac Michaud, 7 months ago

Browse source code at

Authors: Isaac Michaud [cre, aut]

Documentation:   PDF Manual  

GPL-3 license

Imports Rcpp, stats, graphics

Suggests testthat, parallel, tgp

Linking to Rcpp, RcppArmadillo, BH

See at CRAN