The Randomized Dependence Coefficient

Abstract

We introduce the Randomized Dependence Coefficient (RDC), a measure of non-linear dependence between random variables of arbitrary dimension based on the Hirschfeld-Gebelein-Rényi Maximum Correlation Coefficient. RDC is defined in terms of correlation of random non-linear copula projections; it is invariant with respect to marginal distribution transformations, has low computational cost and is easy to implement: just five lines of R code, included at the end of the paper.

References in corpus (3)

• Kernel-based Conditional Independence Test and Application in Causal Discovery
• Rejoinder: Brownian distance covariance
• Copula-based Kernel Dependency Measures

Cited by in corpus (24)

• Stochastic Configuration Networks Ensemble for Large-Scale Data Analytics
• Modern Data Analytics Approach to Predict Creep of High-Temperature Alloys
• Large-Scale Kernel Methods for Independence Testing
• Randomized Nonlinear Component Analysis
• Measuring dependence powerfully and equitably
• On the role of nonlinear correlations in reduced-order modeling
• Kernel Distribution Embeddings: Universal Kernels, Characteristic Kernels and Kernel Metrics on Distributions
• Maximum Relevance and Minimum Redundancy Feature Selection Methods for a Marketing Machine Learning Platform
• DeepDB: Learn from Data, not from Queries!
• Model Order Reduction in Neuroscience
• FLAT: Fast, Lightweight and Accurate Method for Cardinality Estimation
• Understanding Convolutional Neural Networks with Information Theory: An Initial Exploration
• A principle feature analysis
• BayesCard: Revitilizing Bayesian Frameworks for Cardinality Estimation
• Fairness-Aware Neural Réyni Minimization for Continuous Features
• On the Sample Complexity of HGR Maximal Correlation Functions for Large Datasets
• Copula Index for Detecting Dependence and Monotonicity between Stochastic Signals
• Improving Conditional Coverage via Orthogonal Quantile Regression
• Measuring Dependence with Matrix-based Entropy Functional
• FSPN: A New Class of Probabilistic Graphical Model
• An Extreme-Value Approach for Testing the Equality of Large U-Statistic Based Correlation Matrices
• ORCCA: Optimal Randomized Canonical Correlation Analysis
• Kernel Hypothesis Testing with Set-valued Data
• A Note on Simulation-Based Inference by Matching Random Features