Leaky FinFET for Reservoir Computing with Temporal Signal Processing

Abstract

Reservoir computing can greatly reduce the hardware and training costs of recurrent neural networks with temporal data processing. To implement reservoir computing in a hardware form, physical reservoirs transforming sequential inputs into a high-dimensional feature space are necessary. In this work, a physical reservoir with a leaky fin-shaped field-effect transistor (L-FinFET) is demonstrated by the positive use of a short-term memory property arising from the absence of an energy barrier to suppress the tunneling current. Nevertheless, the L-FinFET reservoir does not lose its multiple memory states. The L-FinFET reservoir consumes very low power when encoding temporal inputs because the gate serves as an enabler of the write operation, even in the off-state, due to its physical insulation from the channel. In addition, the small footprint area arising from the scalability of the FinFET due to its multiple-gate structure is advantageous for reducing the chip size. After the experimental proof of 4-bit reservoir operations with 16 states for temporal signal processing, handwritten digits in the Modified National Institute of Standards and Technology dataset are classified by reservoir computing.

High-Dimensional Physical Reservoir with Back-End-of-Line-Compatible Tin Monoxide Thin-Film Transistor

“…Reservoir computing (RC) has gained significant attention as a promising approach for efficient and powerful neuromorphic information processing for pattern recognition and time-series analysis. − Unlike conventional neural networks relying on complex training algorithms, RC utilizes the dynamics of a recurrent dynamical system within the reservoir to achieve remarkable computational capabilities. Physical RC, a growing subfield of RC, explores the implementation of reservoirs using physical devices such as dynamic memristors. − These physical reservoirs capitalize on the devices’ fading memory and nonlinearity characteristics, which map the input data onto a higher-dimensional feature space. Meanwhile, the performance of the RC system is significantly affected by the dimensionality of the reservoirs, which refers to the number of input neurons (or the number of components in the output vector from the reservoir) in the readout network. ,, The RC system can achieve high accuracy by increasing the dimensionality of the reservoirs, enhancing its ability to capture and distinguish information from the input signals.…”