https://doi.org/10.1140/epjs/s11734-023-00773-0
Review
RC-MHM: reservoir computing with a 2D memristive hyperchaotic map
1
College of Artificial Intelligence, Southwest University, Beibei, 400715, Chongqing, China
2
College of Computer and Information Science, Chongqing Normal University, Shapingba, 401331, Chongqing, China
Received:
19
November
2022
Accepted:
24
January
2023
Published online:
2
March
2023
High-performance reservoir computing has to rely on its highly rich internal dynamic. Chaotic systems based on memristors are widely employed in reservoir computing because of their wealthy nonlinear dynamic behaviors, but there are few studies and applications related to discrete memristor models compared to continuous memristor models. This paper proposes a brand new discrete memristor model, and a two-dimensional hyper chaotic maps is constructed by coupling the discrete memristor model and a sinusoidal function. The map has rich nonlinear dynamic behaviors which cater to the need for RC, therefore, we apply this map to enhance the richness of the state of the reservoir, and test the reservoir’s computing capability by the emulation of NARMA and Santa Fe time series prediction task. For this purpose, using the method of the single node with delayed feedback, the proposed reservoir can avoid the selection of meta-parameters, and fully exploits the time dependence of the map states to replace the interactions between the internal nodes. As far as we know, this is the first time that time series analysis using the discrete memristive map, and the experimental results demonstrate that the hyperchaotic map based on the discrete memristor can effectively optimize the performance of the reservoir computing. Moreover, the proposed reservoir shows several attractive features including simple design, ease to calculate, and robustness.
Musha Ji’e, Shengjie Xu and Dengwei Yan have contributed equally to this work.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) 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.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2023. Springer Nature or its licensor (e.g. a society or other partner) 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.
