Current browse context:
cond-mat.stat-mech
Change to browse by:
References & Citations
Condensed Matter > Statistical Mechanics
Title: A generalization of the maximum entropy principle for curved statistical manifolds
(Submitted on 13 May 2021 (v1), last revised 29 Mar 2022 (this version, v2))
Abstract: The maximum entropy principle (MEP) is one of the most prominent methods to investigate and model complex systems. Despite its popularity, the standard form of the MEP can only generate Boltzmann-Gibbs distributions, which are ill-suited for many scenarios of interest. As a principled approach to extend the reach of the MEP, this paper revisits its foundations in information geometry and shows how the geometry of curved statistical manifolds naturally leads to a generalization of the MEP based on the R\'enyi entropy. By establishing a bridge between non-Euclidean geometry and the MEP, our proposal sets a solid foundation for the numerous applications of the R\'enyi entropy, and enables a range of novel methods for complex systems analysis.
Submission history
From: Pablo Morales [view email][v1] Thu, 13 May 2021 12:51:08 GMT (346kb,D)
[v2] Tue, 29 Mar 2022 08:44:37 GMT (351kb,D)
Link back to: arXiv, form interface, contact.