Regensburg 2025 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 5: Poster

AKPIK 5.6: Poster

Donnerstag, 20. März 2025, 15:00–16:30, P2

Latent Measures of Memory and Stochasticity in Dynamical Systems: Murphy’s Law of Tumbling Toast — •Janine Graser, Atreya Majumdar, Kübra Kalkan, Ross Knapman, and Karin Everschor-Sitte — Faculty of Physics and Center for Nanointegration Duisburg-Essen (CENIDE), University of DuisburgEssen, 47057 Duisburg, Germany

Murphy’s Law, which suggests that "anything that can go wrong will go wrong," is often exemplified by toast landing butter-side down. In reality, a toast falling from a table can be described by Newtonian mechanics and is bound to fall on the butter side under standard conditions [1]. Here, the fall is modelled through its seemingly hidden aspects (table height and toast asymmetry because of the butter).

We revisit the tumbling toast problem using the data-driven machine learning tools - latent entropy and latent dimension -introduced by Horenko et al. [2]. We develop a Python-based implementation that characterizes the fall using these latent measures. This approach has broader applications in other dynamical systems, such as predicting and optimizing magnetic material properties.
[1] R. A. J. Matthews, Eur. J. Phys. 16 172 (1995)
[2] I. Horenko et al., Commun. Appl. Math. Comput. Sci. 16 267-297 (2021).

Keywords: Machine Learning; Memory; Stochasticity; Dynamical Systems; Magnetic Materials