Evaluation of the computational capabilities of a memristive random network (MN3) under the context of reservoir computing

Laura E Suarez; Jack D Kendall; Juan C Nino

doi:10.1016/j.neunet.2018.07.003

Evaluation of the computational capabilities of a memristive random network (MN³) under the context of reservoir computing

Neural Netw. 2018 Oct:106:223-236. doi: 10.1016/j.neunet.2018.07.003. Epub 2018 Jul 25.

Authors

Laura E Suarez¹, Jack D Kendall², Juan C Nino³

Affiliations

¹ Department of Industrial Engineering, Universidad de los Andes, Bogotá, Colombia; Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32601, USA. Electronic address: laura.suarez@mail.mcgill.ca.
² Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32601, USA. Electronic address: jackdkendall@ufl.edu.
³ Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32601, USA. Electronic address: jnino@mse.ufl.edu.

PMID: 30077960
DOI: 10.1016/j.neunet.2018.07.003

Abstract

This work presents the simulation results of a novel recurrent, memristive neuromorphic architecture, the MN³ and explores its computational capabilities in the performance of a temporal pattern recognition task by considering the principles of the reservoir computing approach. A simple methodology based on the definitions of ordered and chaotic dynamical systems was used to determine the separation and fading memory properties of the architecture. The results show the potential use of this architecture as a reservoir for the on-line processing of time-varying inputs.

Keywords: Memristive Networks; Neuromorphic Engineering; Reservoir Computing; Speech Recognition.

Publication types

Evaluation Study

MeSH terms

Memory
Nanofibers
Neural Networks, Computer*
Speech Recognition Software* / trends