Fibonacci-Driven Recursive Ensembles: Algorithms, Convergence, and Learning Dynamics Paper III of the Fibonacci Ensemble Trilogy

Ernest Fokoué · 2026

Read Fibonacci-Driven Recursive Ensembles: Algorithms, Convergence, and Learning Dynamics Paper III of the Fibonacci Ensemble Trilogy by Ernest Fokoué (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-01055.txt.

Description

Ernest Fokoué builds a theory of ensemble learning in which each new model leans on the two that came just before it. Most classic methods, such as boosting, add one model at a time with no memory of earlier steps beyond a running sum. The Fibonacci-style update studied here gives the ensemble a memory, so past structure shapes each new step.

The paper sets out a family of update rules, covering Fibonacci, tribonacci, and higher-order versions, plus smooth time limits written as differential equations. Fokoué proves when these flows settle, when they stay stable, and how well they should do on new data. Tests with three classic smoothers show gains over fixed weighting. The work closes a trilogy that grew from Fibonacci weighting into a full theory of learning systems that evolve over time.

Catalog record

CREATORErnest Fokoué
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-01055

Browse the full library · About the Tower of Babel