A chaotic discrete-time continuous-state Hopfield network with piecewise-affine activation functions

DOI10.3934/DCDSB.2023035MaRDI QIDQ6096987zbMATH OpenOpenAlexFDO

Authors Benito Pires

Publication date 9 June 2023

Published in Discrete and Continuous Dynamical Systems. Series B (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2210.00414

zbMATH Keywords

chaotic neural networkHopfield networkCantor attractorpiecewise-affine activation function

Mathematics Subject Classification ID

Dynamical systems involving maps of the interval (37E05) Chaotic behavior of solutions of difference equations (39A33)

Abstract: We construct a chaotic discrete-time continuous-state Hopfield network with piecewise-affine nonnegative activation functions and weight matrix with small positive entries. More precisely, there exists a Cantor set $C$ in the state space such that the network has sensitive dependence on initial conditions at initial states in $C$ and the network orbit of each initial state in $C$ has $C$ as its $omega$-limit set. The approach we use is based on tools developed and employed recently in the study of the topological dynamics of piecewise-contractions. The parameters of the chaotic network are explicitly given.

Recommendations

• Chaos in discrete neural networks with \(n\) neurons
• Global dynamics of discrete neural networks allowing non-monotonic activation functions
• scientific article; zbMATH DE number 6403652
• Chaos of a new class of Hopfield neural networks
• A study of network dynamics

Cites work

• scientific article; zbMATH DE number 1433445 (Why is no real title available?)
• A generalized convergence theorem for neural networks
• A switched server system semiconjugate to a minimal interval exchange
• Asymptotically periodic piecewise contractions of the interval
• Dynamics of piecewise contractions of the interval
• Energy functions in neural networks with continuous local functions
• Neural networks and physical systems with emergent collective computational abilities
• Neurons with graded response have collective computational properties like those of two-state neurons
• Symbolic dynamics of piecewise contractions
• Topological dynamics of piecewise \(\lambda\)-affine maps

This page was built for publication: A chaotic discrete-time continuous-state Hopfield network with piecewise-affine activation functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6096987)