Research map / Probabilistic and scientific modeling / Molecules and proteins
SHGR: A Generalized Maximal Correlation Coefficient
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes SHGR, a generalized maximal correlation coefficient extending the Hirschfeld–Gebelein–Rényi (HGR) measure to capture nonlinear and multivariate dependencies.
Problem
Traditional correlation measures (Pearson, Spearman) miss nonlinear and multivariate dependencies, and estimating the HGR maximal correlation is difficult.
Approach
Generalizes the HGR maximal correlation, which seeks the highest Pearson correlation achievable via nonlinear transformations of two variables, into a coefficient (SHGR) intended to handle multivariate dependence; the abstract is truncated so the precise estimator is not stated.
Experimental setup
Builds on: HGR maximal correlation
Closest papers by meaning
- Adjust Pearson's $r$ to Measure Arbitrary Monotone Dependence (NeurIPS 2024 · similarity 0.83)
- Learning from higher-order correlations, efficiently: hypothesis tests, random features, and neural networks (NeurIPS 2024 · similarity 0.83)
- Jacobian Regularizer-based Neural Granger Causality (ICML 2024 · similarity 0.82)
- On Sparse Canonical Correlation Analysis (NeurIPS 2024 · similarity 0.82)
- Sliding Down the Stairs: How Correlated Latent Variables Accelerate Learning with Neural Networks (ICML 2024 · similarity 0.82)
- TIC-TAC: A Framework For Improved Covariance Estimation In Deep Heteroscedastic Regression (ICML 2024 · similarity 0.81)
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.