On Unstable Fixed Points in Modern Continuous Hopfield Networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autor principal: Beise, Hans-Peter
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915913188507648
author Beise, Hans-Peter
author_facet Beise, Hans-Peter
contents The recently introduced continuous Hopfield network (see Ramsauer et al.) exhibits large memorization capabilities, which manifest as attractive fixed points of its update rule -- a differentiable function consisting of two linear mappings composed with the scaled softmax function. The authors of the aforementioned work provide proofs for the existence and approximate position of such attractive fixed points. For the softmax function alone, the fixed point structure has been fully characterized in earlier work by P. Tiňo, from which it turns out that for sufficiently large scaling factors there are exponentially more unstable fixed points than attractive ones. In this work, we complement the findings of Ramsauer et al. by showing that, under natural geometric conditions on the vectors defining the continuous Hopfield network, unstable fixed points must occur, analogous to the findings of Tiňo. Our results show that, under these geometric conditions, continuous Hopfield networks necessarily admit additional unstable fixed points associated with higher-dimensional faces of the pattern polytope.
format Preprint
id arxiv_https___arxiv_org_abs_2603_27804
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle On Unstable Fixed Points in Modern Continuous Hopfield Networks
Beise, Hans-Peter
Dynamical Systems
37B35, 37N99
The recently introduced continuous Hopfield network (see Ramsauer et al.) exhibits large memorization capabilities, which manifest as attractive fixed points of its update rule -- a differentiable function consisting of two linear mappings composed with the scaled softmax function. The authors of the aforementioned work provide proofs for the existence and approximate position of such attractive fixed points. For the softmax function alone, the fixed point structure has been fully characterized in earlier work by P. Tiňo, from which it turns out that for sufficiently large scaling factors there are exponentially more unstable fixed points than attractive ones. In this work, we complement the findings of Ramsauer et al. by showing that, under natural geometric conditions on the vectors defining the continuous Hopfield network, unstable fixed points must occur, analogous to the findings of Tiňo. Our results show that, under these geometric conditions, continuous Hopfield networks necessarily admit additional unstable fixed points associated with higher-dimensional faces of the pattern polytope.
title On Unstable Fixed Points in Modern Continuous Hopfield Networks
topic Dynamical Systems
37B35, 37N99
url https://arxiv.org/abs/2603.27804