Reservoir computing using dynamic memristors for temporal information processing

Chao Du,Mohammed A Zidan,Fuxi Cai,Wei D Lu,Wen Ma,Seung Hwan Lee

doi:10.1038/s41467-017-02337-y

Abstract

Reservoir computing systems utilize dynamic reservoirs having short-term memory to project features from the temporal inputs into a high-dimensional feature space. A readout function layer can then effectively analyze the projected features for tasks, such as classification and time-series analysis. The system can efficiently compute complex and temporal data with low-training cost, since only the readout function needs to be trained. Here we experimentally implement a reservoir computing system using a dynamic memristor array. We show that the internal ionic dynamic processes of memristors allow the memristor-based reservoir to directly process information in the temporal domain, and demonstrate that even a small hardware system with only 88 memristors can already be used for tasks, such as handwritten digit recognition. The system is also used to experimentally solve a second-order nonlinear task, and can successfully predict the expected output without knowing the form of the original dynamic transfer function.

Highlights

Reservoir computing systems utilize dynamic reservoirs having short-term memory to project features from the temporal inputs into a high-dimensional feature space
The reservoir itself must have short-term memory. It has been mathematically shown[3] that an Reservoir computing (RC) system only needs to possess two very unrestrictive properties to achieve universal computation power for time-varying inputs: point-wise separation property for the reservoir, which means that all output-relevant differences in the input series u1(·) and u2(·) before time t are reflected in the corresponding reservoir internal states x1(·) and x2(·) that are separable; and approximation property for the readout function, which means that the readout function can map the current reservoir state to the desired current output with required accuracy
We demonstrate a memristor-based RC system by utilizing the internal, short-term ionic dynamics of memristor devices

Summary

Results

Memristors are two terminal resistive elements with memory effects, where the state of the device depends on one or more internal state variables and can be modulated depending on the history of external stimulation[16,17,18,19]. At the single-device level, memristors have been shown to be able to emulate synaptic functions by storing analog synaptic weights and modulate the connection strength between the input and output neurons[22,23,24], while recent studies have demonstrated that these devices can even emulate synaptic effects faithfully based on internal ionic dynamics[13,14,15,27,28]. Memristor devices with short-term memory effects[13,14,15] are used in this study to act as the reservoir in an RC system. The switching layer of the WOx based device was memory (i.e., volatile) designed to exhibit short-term behavior[13,14,15]

30 Cell number

Discussion

Methods