Home > Publications database > Phase Space Analysis of Chaotic Neural Networks |
Preprint | FZJ-2022-03707 |
; ; ; ;
2022
arXiv
This record in other databases:
Please use a persistent id in citations: http://hdl.handle.net/2128/32902 doi:10.48550/ARXIV.2210.07877
Abstract: We analytically determine the distribution of fixed points in a canonical model of a chaotic neural network. This distribution reveals that fixed points and dynamics are confined to separate shells in phase space. Furthermore, the distribution enables us to determine the eigenvalue spectra of the Jacobian at the fixed points. Perhaps counter-intuitively, the velocity of the dynamics is strongly correlated with the direction imposed by the nearest fixed point despite the spatial separation. We propose that this influence of the fixed points is mediated by tangentially fixed lines.
Keyword(s): Disordered Systems and Neural Networks (cond-mat.dis-nn) ; Chaotic Dynamics (nlin.CD) ; FOS: Physical sciences
![]() |
The record appears in these collections: |