Paper Info
Title
Coexisting Behaviors of Asymmetric Attractors in Hyperbolic-Type Memristor based Hopfield Neural Network.
Abstract
A new hyperbolic-type memristor emulator is presented and its frequency-dependent pinched hysteresis loops are analyzed by numerical simulations and confirmed by hardware experiments. Based on the emulator, a novel hyperbolic-type memristor based 3-neuron Hopfield neural network (HNN) is proposed, which is achieved through substituting one coupling-connection weight with a memristive synaptic weight. It is numerically shown that the memristive HNN has a dynamical transition from chaotic, to periodic, and further to stable point behaviors with the variations of the memristor inner parameter, implying the stabilization effect of the hyperbolic-type memristor on the chaotic HNN. Of particular interest, it should be highly stressed that for different memristor inner parameters, different coexisting behaviors of asymmetric attractors are emerged under different initial conditions, leading to the existence of multistable oscillation states in the memristive HNN. Furthermore, by using commercial discrete components, a nonlinear circuit is designed and PSPICE circuit simulations and hardware experiments are performed. The results simulated and captured from the realization circuit are consistent with numerical simulations, which well verify the facticity of coexisting asymmetric attractors' behaviors.
Year
DOI
Venue
2017
10.3389/fncom.2017.00081
FRONTIERS IN COMPUTATIONAL NEUROSCIENCE
Keywords
Field
DocType
hyperbolic-type memristor,Hopfield neural network (HNN),asymmetric attractors,coexisting behaviors,hardware experiment
Attractor,Oscillation,Nonlinear system,Computer science,Control theory,Artificial intelligence,Chaotic,Artificial neural network,Topology,Memristor,Synaptic weight,Periodic graph (geometry),Machine learning
Journal
Volume
ISSN
Citations 
11
1662-5188
6
PageRank 
References 
Authors
0.58
15
6
Name
Order
Citations
PageRank
Bocheng Bao111919.50
Hui Qian261.26
Quan Xu3287.13
Mo Chen4298.53
jiang wang5113.38
Yajuan Yu6163.26