https://doi.org/10.1140/epjs/s11734-022-00642-2
Regular Article
ReLU-type memristor-based Hopfield neural network
School of Electrical and Automation Engineering, School of Computer Science and Technology, Nanjing Normal University, 210023, Nanjing, China
Received:
2
March
2022
Accepted:
15
July
2022
Published online:
4
August
2022
Due to the simple circuit realization, this paper proposes a ReLU-type memristor emulator firstly, whose pinched hysteresis loops are analyzed via numerical measures and certified via circuit simulations. On account of this emulator, a novel ReLU-type memristor-based Hopfield neural network (HNN) is presented, which is acquired by replacing a resistive interconnection synaptic weight with a memristive synaptic weight. The memristive HNN model has line equilibrium, and its stability is always unstable for different memristor coupling intensions. Furthermore, utilizing several numerical measures like bifurcation plots, mean value diagrams, phase portraits, and time sequences, we confirm that the ReLU-type memristor-based HNN model behaves the coexistence of multi-stable patterns of the double-scroll chaotic patterns with diverse topologies and periodic patterns with diverse topologies and periodicities. Of great interest, we demonstrate that transition behaviors and memristor initial boosting behaviors are also emerged in such memristive HNN model. Finally, the facticity of intricate kinetics is effectively validated by analog circuit simulations.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2022. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
