Neuro-Skins: Dynamics, Plasticity and Effect of Neuron Type and Cell Size on Their Response

We are introducing a new type of membrane, called neuro-skin or neuro-membrane. It is comprised of neurons embedded in a plastic membrane. The skin is smart and adaptive and is capable of providing desirable response to inputs intelligently. This way, the neuro-skin can be considered as a new type of neural network with adaptivity and learning capabilities. However, in this paper, only the response of neuro-skins to a dynamic input is studied. The membrane is modelled by nonlinear dynamic finite elements. Each finite element is considered as a cell of the neuro-skin which has a neuron. The neuron is the intelligent nucleus of the element. So, the finite elements are called finite neuro-elements (FNEs). Each FNE receives feedback excitation from its own neuron, as well as from its neighbouring neurons. Contrary to dynamic plastic continuous neural networks previously studied by the authors, the neurons in a neuro-skin do not apply concentrated loads but they apply traction stresses to the surface of NFEs. The membrane is in fact a skin made up of intelligent cells representing both neural activity and mechanical plasticity. The effect of neuron type and cell size on the response of neuro-skins is studied. Trainability is another issue which is not discussed in this paper. We have used the terms neuro-skin and neuro-membrane interchangeably.

[2]  Jinde Cao,et al.  Exponential Synchronization of Coupled Stochastic Memristor-Based Neural Networks With Time-Varying Probabilistic Delay Coupling and Impulsive Delay , 2016, IEEE Transactions on Neural Networks and Learning Systems.

[3]  Stability analysis on a class of nonlinear continuous neural networks , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[4]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[5]  V. Fromion.  Lipschitz continuous neural networks on L/sub p/ , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[6]  N.K. Sinha,et al.  Dynamic neural networks: an overview , 2000, Proceedings of IEEE International Conference on Industrial Technology 2000 (IEEE Cat. No.00TH8482).

[7]  Abdolreza Joghataie,et al.  Nonlinear Analysis of Concrete Gravity Dams by Neural Networks , 2009 .

[8]  Gaetan Libert,et al.  Dynamic recurrent neural networks: a dynamical analysis , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[9]  Anil K. Chopra,et al.  Dynamics of Structures: Theory and Applications to Earthquake Engineering , 1995 .

[10]  Jinde Cao,et al.  Synchronization of fractional-order complex-valued neural networks with time delay , 2016, Neural Networks.

[11]  Jasbir S. Arora,et al.  Introduction to Optimum Design , 1988 .

[12]  Abdolreza Joghataie,et al.  Neuroplasticity in dynamic neural networks comprised of neurons attached to adaptive base plate , 2016, Neural Networks.

[13]  Yang Gao,et al.  Oscillation propagation in neural networks with different topologies. , 2011, Physical review. E, Statistical, nonlinear, and soft matter physics.

[14]  A. Margaris,et al.  A MUTUAL INFORMATION-BASED METHOD FOR THE ESTIMATION OF THE DIMENSION OF CHAOTIC DYNAMICAL SYSTEMS USING NEURAL NETWORKS , 2008 .

[15]  Abdolreza Joghataie,et al.  Dynamic Analysis of Nonlinear Frames by Prandtl Neural Networks , 2008 .

[16]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[17]  Abdolreza Joghataie,et al.  Simulating Dynamic Plastic Continuous Neural Networks by Finite Elements , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[18]  Kurt Hornik,et al.  FEED FORWARD NETWORKS ARE UNIVERSAL APPROXIMATORS , 1989 .

[19]  Abdolreza Joghataie,et al.  Transforming Results from Model to Prototype of Concrete Gravity Dams Using Neural Networks , 2011 .

[20]  Yu Qian,et al.  Pattern formation in oscillatory complex networks consisting of excitable nodes. , 2010, Physical review. E, Statistical, nonlinear, and soft matter physics.

[21]  Naoki Masuda,et al.  Self-organization of feed-forward structure and entrainment in excitatory neural networks with spike-timing-dependent plasticity. , 2008, Physical review. E, Statistical, nonlinear, and soft matter physics.

[22]  Jinde Cao,et al.  Existence and Globally Asymptotic Stability of Equilibrium Solution for Fractional-Order Hybrid BAM Neural Networks with Distributed Delays and Impulses , 2017, Complex..

[23]  K. Bathe.  Finite Element Procedures , 1995 .

[24]  L. Li,et al.  Dynamics of some neural network models with delay. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[25]  M. Brokate,et al.  Hysteresis and Phase Transitions , 1996 .

[26]  Jinde Cao,et al.  Adaptive synchronization of fractional-order memristor-based neural networks with time delay , 2015, Nonlinear Dynamics.

[27]  Jinde Cao,et al.  Robust fixed-time synchronization for uncertain complex-valued neural networks with discontinuous activation functions , 2017, Neural Networks.

[28]  Jinde Cao,et al.  Fixed-time synchronization of delayed memristor-based recurrent neural networks , 2017, Science China Information Sciences.

[29]  Satish S. Udpa,et al.  Finite-element neural networks for solving differential equations , 2005, IEEE Transactions on Neural Networks.

[30]  Allon Guez,et al.  On the stability, storage capacity, and design of nonlinear continuous neural networks , 1988, IEEE Trans. Syst. Man Cybern..

[31]  Jacek M. Zurada,et al.  Numerical modeling of continuous-time fully coupled neural networks , 1991, [Proceedings] 1991 IEEE International Joint Conference on Neural Networks.

[32]  Guy Littlefair,et al.  Application of FE-based neural networks to dynamic problems , 1999, ICONIP'99. ANZIIS'99 & ANNES'99 & ACNN'99. 6th International Conference on Neural Information Processing. Proceedings (Cat. No.99EX378).

[33]  Shun-ichi Amari,et al.  Mathematical foundations of neurocomputing , 1990, Proc. IEEE.

[34]  Mojtaba Farrokh,et al.  Adaptive simulation of hysteresis using neuro-Madelung model , 2016 .

[35]  N. Petrinic,et al.  Introduction to computational plasticity , 2005 .

[36]  Abdolreza Joghataie,et al.  Designing High-Precision Fast Nonlinear Dam Neuro-Modelers and Comparison with Finite-Element Analysis , 2013 .

[37]  Jinde Cao,et al.  Nonlinear Measure Approach for the Stability Analysis of Complex-Valued Neural Networks , 2015, Neural Processing Letters.

[38]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[39]  Abdolreza Joghataie,et al.  Modeling Hysteretic Deteriorating Behavior Using Generalized Prandtl Neural Network , 2015 .

[40]  V. Fromion.  Lipschitz continuous neural networks on L , 2000 .

[41]  Jinde Cao,et al.  Delay-Independent Stability of Riemann–Liouville Fractional Neutral-Type Delayed Neural Networks , 2017, Neural Processing Letters.

[42]  Stephen Grossberg,et al.  Absolute stability of global pattern formation and parallel memory storage by competitive neural networks , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[43]  Jinde Cao,et al.  Dynamics in fractional-order neural networks , 2014, Neurocomputing.