Design of Multilayer Perceptron Network Based on Metal-Oxide Memristive Devices

A key problem at hardware implementation of artificial neural networks based on memristors (ANNM) is to ensure the required accuracy of their operation at the transition from models to real fabricated memristive devices. Due to a number of factors, such as the imperfections in stateof- the-art memristors and memristive arrays, ANNM design and tuning methods, additional computation errors occur during the process of ANNM hardware implementation. The article proposes a general approach to the simulation and design of a multilayer perceptron (MLP) network implemented with original cross-bar arrays of metal-oxide memristive devices. The proposed approach is based on the theory of engineering tolerances, simulation and the design of experiments. The authors present the research results for the ANNM trained to solve the problem of nonlinear classification for a bidirectional adaptive neural interface.

[1]  Alexander I. Galushkin,et al.  Neural Networks Theory , 2007 .

[2]  Dmitri B. Strukov,et al.  Towards the Development of Analog Neuromorphic Chip Prototype with 2.4M Integrated Memristors , 2019, 2019 IEEE International Symposium on Circuits and Systems (ISCAS).

[3]  A. I. Galushkin,et al.  The research of memristor-based neural network components operation accuracy in control and communication systems , 2015, 2015 International Siberian Conference on Control and Communications (SIBCON).

[4]  Leon O. Chua,et al.  Φ memristor: Real memristor found , 2019, Journal of Applied Physics.

[5]  S.N. Danilin,et al.  Algorithm for Determining Optimum Operation Tolerances of Memristor-Based Artificial Neural Networks , 2017, 2017 IVth International Conference on Engineering and Telecommunication (EnT).

[6]  Victor B. Kazantsev,et al.  One-Board Design and Simulation of Double-Layer Perceptron Based on Metal-Oxide Memristive Nanostructures , 2018, IEEE Transactions on Emerging Topics in Computational Intelligence.

[7]  Catherine D. Schuman,et al.  A Survey of Neuromorphic Computing and Neural Networks in Hardware , 2017, ArXiv.

[8]  Max Talanov,et al.  The memristive artificial neuron high level architecture for biologically inspired robotic systems , 2017, 2017 International Conference on Mechanical, System and Control Engineering (ICMSC).

[9]  K. E. Nikiruy,et al.  Adaptive Properties of Spiking Neuromorphic Networks with Synapses Based on Memristive Elements , 2019 .

[10]  Leon O. Chua,et al.  Five non-volatile memristor enigmas solved , 2018, Applied Physics A.

[11]  Dmitri B. Strukov,et al.  Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits , 2017, Nature Communications.

[12]  S.N. Danilin,et al.  Determining Operation Tolerances of Memristor-Based Artificial Neural Networks , 2016, 2016 International Conference on Engineering and Telecommunication (EnT).

[13]  A. V. Emelyanov,et al.  Spike-timing-dependent plasticity of polyaniline-based memristive element , 2018 .

[14]  Leon O. Chua,et al.  A Circuit-Based Neural Network with Hybrid Learning of Backpropagation and Random Weight Change Algorithms , 2016, Sensors.