New NMINE neural estimator improves mutual information measurement

Researchers have developed a novel neural estimator called NMINE for normalized mutual information, designed to measure statistical dependence between continuous and multidimensional variables. This approach combines a MINE-based estimator for mutual information with MI-NEE-inspired estimators for marginal entropies, utilizing the Donsker-Varadhan representation. Experiments indicate that NMINE offers improved accuracy over existing methods, particularly in higher dimensions, suggesting its potential for dependency measurement in complex continuous settings. AI

IMPACT This new neural estimator could enhance dependency measurement in complex continuous multidimensional data, potentially benefiting machine learning applications.

RANK_REASON The cluster contains a research paper detailing a new method for estimating normalized mutual information. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv:1801.04062
• arXiv:1905.12957
• arXiv:2405.04980
• arXiv:2409.16768
• arXiv:2410.07642
• Mutual Information Neural Estimation
• NMINE
• Petra Eerikinharju

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →