Abstract:Abstract Reservoir computing (RC) offers efficient temporal information processing with low training cost. All-ferroelectric implementation of RC is appealing because it can fully exploit the merits of ferroelectric memristors (e.g., good controllability); however, this has been undemonstrated due to the challenge of developing ferroelectric memristors with distinctly different switching characteristics specific to the reservoir and readout network. Here, we experimentally demonstrate an all-ferroelectric RC system whose reservoir and readout network are implemented with volatile and nonvolatile ferroelectric diodes (FDs), respectively. The volatile and nonvolatile FDs are derived from the same Pt/BiFeO 3 /SrRuO 3 structure via the manipulation of an imprint field ( E imp ). It is shown that the volatile FD with E imp exhibits short-term memory and nonlinearity while the nonvolatile FD with negligible E imp displays long-term potentiation/depression, fulfilling the functional requirements of the reservoir and readout network, respectively. Hence, the all-ferroelectric RC system is competent for handling various temporal tasks. In particular, it achieves an ultralow normalized root mean square error of 0.017 in the Hénon map time-series prediction. Besides, both the volatile and nonvolatile FDs demonstrate long-term stability in ambient air, high endurance, and low power consumption, promising the all-ferroelectric RC system as a reliable and low-power neuromorphic hardware for temporal information processing.

What problem does this paper attempt to address?

The paper mainly aims to address the following issues: ### Research Background and Objectives - **Background**: Reservoir Computing (RC) is an efficient method for processing temporal information with low training costs, suitable for handling dynamic temporal tasks. However, in hardware-implemented RC systems, the reservoir and readout network require memory devices with different switching characteristics (volatile and non-volatile), which is a challenge in current technology. - **Objective**: The research aims to experimentally demonstrate a reservoir computing system implemented entirely with ferroelectric materials, where the reservoir and readout network are composed of ferroelectric diodes (FD) with volatile and non-volatile characteristics, respectively. ### Key Issues Addressed - **Development of New Ferroelectric Devices**: By regulating the properties of ferroelectric materials, design ferroelectric diodes that exhibit both volatile and non-volatile switching behaviors for constructing the reservoir and readout network. - **Integrated System Verification**: Integrate these new ferroelectric devices into a complete RC system and verify their performance on various temporal tasks, including curve discrimination, digit recognition, waveform classification, and Hénon map prediction. - **Improving Reliability and Energy Efficiency**: Demonstrate the advantages of ferroelectric material-based RC systems in terms of environmental stability, durability, and low power consumption to prove their potential as reliable neuromorphic hardware. ### Technical Solutions - **Volatile Ferroelectric Diodes**: By introducing a specific imprint field (E_imp), the ferroelectric diodes exhibit short-term memory effects and nonlinear behavior, suitable for constructing the reservoir. - **Non-Volatile Ferroelectric Diodes**: Without a significant imprint field, they exhibit long-term stability and non-volatile switching characteristics, suitable for use as the readout network. - **Device Structure**: Both types of ferroelectric diodes are based on the Pt/BiFeO3/SrRuO3 structure, but their volatile or non-volatile characteristics are regulated by changing the growth conditions of the BiFeO3 film. - **Experimental Verification**: The different characteristics of the two types of ferroelectric diodes were verified through pulse writing methods, and a complete ferroelectric RC system was constructed, successfully completing multiple tasks. ### Main Contributions - **Theoretical Innovation**: Propose a new approach to reservoir computing by utilizing the different properties of ferroelectric materials. - **Experimental Validation**: Experimentally demonstrate the feasibility of an RC system based on ferroelectric diodes and its high performance in complex temporal tasks. - **Application Prospects**: Prove that ferroelectric material-based RC systems have advantages such as high reliability and low power consumption, providing a new direction for the development of future neuromorphic computing hardware.

All-ferroelectric implementation of reservoir computing

Switching Dynamics in Anti‐Ferroelectric Transistor for Multimodal Reservoir Computing

Physical reservoirs based on MoS 2 –HZO integrated ferroelectric field-effect transistors for reservoir computing systems

Fully Ferroelectric-FETs Reservoir Computing Network for Temporal and Random Signal Processing

Cost‐Effective and Fully Hardware‐Oriented Reservoir Computing Based on IGZO/HZO Ferroelectric Thin‐Film Transistor with Electrically and Optically Distinguishable States

Multilayer Reservoir Computing Based on Ferroelectric Α‐in2se3 for Hierarchical Information Processing

Multilayer Reservoir Computing Based on Ferroelectric α-In2 Se3 for Hierarchical Information Processing

Analog reservoir computing via ferroelectric mixed phase boundary transistors

Multilayer Reservoir Computing Based on Ferroelectric α‐In <sub>2</sub> Se <sub>3</sub> for Hierarchical Information Processing

Dynamic Memristor-based Reservoir Computing for High-Efficiency Spatiotemporal Signal Processing

A FeFET-based Time-Domain Associative Memory for Multi-bit Similarity Computation

Nonvolatile Ferroelectric Field-Effect Transistors.

Analysis and Fully Memristor-based Reservoir Computing for Temporal Data Classification

Computing-In-Memory Using Ferroelectrics: From Single- to Multi-Input Logic

Reservoir computing based on electric-double-layer coupled InGaZnO artificial synapse

Enhanced Ferroelectricity of Hf‐Based Memcapacitors by Adopting Ti Insert‐Layer and C–V Measurement for Constructing Energy‐Efficient Reservoir Computing Network

A ferroelectric fin diode for robust non-volatile memory

Functional Materials for Memristor‐Based Reservoir Computing: Dynamics and Applications

Ultralow-power in-memory computing based on ferroelectric memcapacitor network

Reconfigurable Physical Reservoir Enabled by Polarization of Ferroelectric Polymer P(VDF–TrFE) and Interface Charge‐Trapping/Detrapping in Dual‐Gate IGZO Transistor

Reservoir computing using dynamic memristors for temporal information processing