AI & ML
A New Perspective on Cumulative Residual R\'enyi Entropy
arXiv stat.TH0 views
arXiv:2610.19903v1 Announce Type: new Abstract: This paper introduces a new cumulative residual R\'enyi entropy (CRRE), which addresses limitations of the existing cumulative residual R\'enyi entropy. The conventional measure suffers from two primary drawbacks: its failure to converge to the cumulative residual entropy (CRE) as the order parameter approaches unity, and its lack of guaranteed positivity. Our proposed CRRE resolves these issues, offering a theoretically consistent and always nonnegative measure of uncertainty in residual lifetimes. We explore its mathematical properties, including its relationship with existing information measures. Furthermore, the concept is extended to its dynamic version, and its time-dependent characteristics are investigated. A non-parametric estimator of the proposed CRRE is developed, its almost sure convergence is established, and is validated through Monte Carlo simulation. The effectiveness of CRRE as a tool for quantifying uncertainty is confirmed through its application in chaotic theory and financial data.
Source:https://arxiv.org/abs/2610.19903PreviousSoftplus Affine(1, 1) Processes: Enlarged Stationarity Region, Stationary Unit Roots, and InferenceNextDrift parameter estimation for interacting particle systems with singular repulsion
Related posts
Scrape LinkedIn job postings without logging in (Python + Google Sheets)
0dev.to
Same model, same defences: 7 document attacks down to 0, chat attacks barely moved
0dev.to
Skill to Create Product Videos
0Hacker News
SOML – A Config Format for Humans
0Hacker News
Egoless Mathematics in the AI Era
0Hacker News
LifeVal
0OpenAI Blog