EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Generalized maximum entropy: when and why you need it

Giuseppe M Ferro iD, Andrea Somazzi, Edwin Pos

Royal Society Open Science · 04.11.2026 · 13 (11)

DOI: 10.1098/rsos.260003 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Royal Society Open Science, online erschienen am 04.11.2026. Verfasst von Giuseppe M Ferro und 2 weiteren Autor:innen. Beteiligte Einrichtung: Center for Statistics and Machine Learning, Princeton University, Princeton, NJ, USA. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The classical maximum entropy principle based on Shannon entropy is widely used to construct least-biased probability distributions from partial information. However, the Shore–Johnson axioms that single out the Shannon functional hinge on strong system independence—an assumption often violated in real-world, strongly correlated systems. When that assumption fails, enforcing the Shannon form imposes an independence the data do not support, and can systematically distort the inferred distribution. We provide a self-contained guide to when and why practitioners should abandon the Shannon form in favour of the one-parameter Uffink–Jizba–Korbel (UJK) family of generalized entropies. After reviewing the Shore–Johnson axioms from an applied perspective, we recall the most commonly used entropy functionals and locate them within the UJK family. The need for generalized entropies is illustrated with two applications, one in economics and the other in ecology. A simple mathematical model worked out in detail further shows the power of generalized maximum-entropy approaches in cases where strong system independence does not hold. We conclude with practical guidelines for choosing an entropy measure and reporting results so that analyses remain transparent and reproducible.

Schlagwörter

Stichwörter der Autor:innen

complex systemsmaximum entropy principlestatistical inference

Autor:innen und Institutionen

  • Center for Statistics and Machine Learning, Princeton University, Princeton, NJ, USA.
  • Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ, USA.
  • IMT School for Advanced Studies Lucca, Lucca, Italy.
  • Quantitative Biodiversity Dynamics, Ecology and Biodiversity, Utrecht University, Utrecht, The Netherlands.
  • Botanic Gardens, Utrecht University, Utrecht, The Netherlands.

ORCID-Profile: Giuseppe M Ferro

Zitieren

APA

Ferro, G. M., Somazzi, A., & Pos, E. (2026). Generalized maximum entropy: when and why you need it. Royal Society Open Science, 13(11). https://doi.org/10.1098/rsos.260003

Vancouver

Ferro GM, Somazzi A, Pos E. Generalized maximum entropy: when and why you need it. Royal Society Open Science. 2026;13(11). doi:10.1098/rsos.260003

BibTeX
@article{Ferro2026,
  author = {Ferro, Giuseppe M and Somazzi, Andrea and Pos, Edwin},
  title = {Generalized maximum entropy: when and why you need it},
  journal = {Royal Society Open Science},
  year = {2026},
  volume = {13},
  number = {11},
  doi = {10.1098/rsos.260003}
}
Kostenlos starten →