A 2D Hyperchaotic Discrete Memristive Map and Application in Reservoir Computing

Compared with continuous memristor, discrete memristor has not been deeply studied. In this work, a new discrete memristor model and its pinched hysteresis loops are explored. Based on this model, a simple 2D hyperchaotic map and its dynamics are exhibited. Reservoir computing is an extension of neural networks, which has been received adequate attention. So far, the application of discrete memristor on reservoir has not been reported. To this end, we consider the discrete memristor-based map as a reservoir and verify its performance by a nonlinear regression task. The results indicate that the memristor-based map can be used effectively as a reservoir and enable reservoir computing systems go further on the way of efficient processing temporal signals in the future.