Abstract:On the basis of Hindmarsh-Rose neuron model, dynamical robustness and the transition of firing modes of multilayer memristive neural network consisting of nonidentical neurons have been investigated in detail. The dynamic effects of memristive synapses coupling which are either cubic order flux or quadratic flux are detected. Our results suggest that for the cubic order flux-controlled memristors increasing electromagnetic induction parameter and parameter of memristor show the tendency to spoil the dynamical robustness of the multilayer memristive neural network. For quadratic flux-controlled memristors, weak electromagnetic induction can spoil the dynamical robustness while strong electromagnetic induction and parameter of memristor have little influence on dynamical robustness. We found that the ratio of inactive neurons switches the firing patterns of the active neuron among periodic bursting, chaotic bursting and spiking-like. In the case of memristive synapse coupling by cubic order flux-controlled memristors, chaotic bursting alternates with period bursting as the ratio of inactive neurons is increased, and the chaotic bursting occupy a large range of the ratio of inactive neurons. In the case of memristive synapse coupling by quadratic flux-controlled memristors, the main firing pattern of the active neuron is chaotic bursting. The distribution of interspike interval can be broadly classified into two groups: those that are long interspike intervals and those that are short interspike intervals. The dynamics of multilayer memristive neural network is also verified in the circuits built on Multisim. These results could give new mechanism explanation for aging transition by applying electromagnetic field to neural network. (c) 2021 Elsevier Inc. All rights reserved.

Influences of Perturbations of Training Pattern Pairs on Fuzzy Morphological Neural Networks

Influences and Control of Perturbations of Training Pattern Pairs on Fuzzy Bidirectional Associative Memories

Influences of Perturbations of Training Pattern Pairs on Max-Product Fuzzy Associative Memory

Memorizing morph patterns in small-world neuronal network

Research on Properties of Max-Product Fuzzy Associative Memory Networks

A Unified Framework Of Morphological Associative Memories

Effect of Similarity Between Patterns in Associative Memory

Synchronization Control of Memristor-Based Recurrent Neural Networks with Perturbations.

Training Artificial Neural Networks by Generalized Likelihood Ratio Method: an Effective Way to Improve Robustness.

A Family of New Morphological Associative Memories

Tolerance of Intrinsic Device Variation in Fuzzy Restricted Boltzmann Machine Network Based on Memristive Nano-Synapses

An Empirical Study on the Effect of Training Data Perturbations on Neural Network Robustness

Dynamical Robustness and Firing Modes in Multilayer Memristive Neural Networks of Nonidentical Neurons.

Coexistence of Cyclic Sequential Pattern Recognition and Associative Memory in Neural Networks by Attractor Mechanisms

PSSAT: A Perturbed Semantic Structure Awareness Transferring Method for Perturbation-Robust Slot Filling.

Perturbation of Spike Timing Benefits Neural Network Performance on Similarity Search

Unified approach for applications of oscillatory associative-memory networks with error-free retrieval

Learning, Storing, and Disentangling Correlated Patterns in Neural Networks.

A Fuzzy Quantum Neural Network and Its Application in Pattern Recognition

Learning in Memristive Neural Networks with Various Connection Patterns

Neural Pattern Similarity in the Left IFG and Fusiform is Associated with Novel Word Learning