Li Lingqi, Xiang Heng, Zheng Haofei, Chien Yu-Chieh, Duong Ngoc Thanh, Gao Jing, Ang Kah-Wee
Department of Electrical and Computer Engineering, National University of Singapore, 4 Engineering Drive 3, Singapore, 117583.
Nanoscale Horiz. 2024 Apr 29;9(5):752-763. doi: 10.1039/d3nh00524k.
Reservoir computing (RC), a variant of recurrent neural networks (RNNs), is well-known for its reduced energy consumption through exclusive focus on training the output weight and its superior performance in handling spatiotemporal information. Implementing these networks in hardware requires devices with superior fading memory behavior. Unlike filament-based two-terminal devices, those relying on ferroelectric switching demonstrate improved voltage reliability, while three-terminal transistors provide additional active control. HfO-based ferroelectric materials such as HfZrO (HZO), have garnered attention for their scalability and seamless integration with CMOS technology. This study implements a RC hardware based on MoS-HZO integrated device structure with enhanced spontaneous polarization field. By adjusting the oxygen vacancy concentration, the devices exhibit consistent responses to both identical and nonidentical voltages, making them suitable for diverse RC applications. The high accuracy of MNIST handwritten digits recognition highlights the rich reservoir states of the traditional RC architecture. Additionally, the impact of masks on RC implementation is assessed, showcasing the device's capability for spatiotemporal signal analysis. This development paves the way for implementing energy-efficient and high-performance computing solutions.
储层计算（RC）是循环神经网络（RNN）的一种变体，因其专注于训练输出权重从而降低能耗以及在处理时空信息方面的卓越性能而闻名。在硬件中实现这些网络需要具有卓越衰退记忆行为的器件。与基于细丝的两终端器件不同，那些依赖铁电开关的器件表现出更高的电压可靠性，而三终端晶体管则提供额外的有源控制。诸如HfZrO（HZO）等基于HfO的铁电材料因其可扩展性以及与CMOS技术的无缝集成而受到关注。本研究基于具有增强自发极化场的MoS-HZO集成器件结构实现了一种RC硬件。通过调整氧空位浓度，这些器件对相同和不同电压均表现出一致的响应，使其适用于各种RC应用。MNIST手写数字识别中的高精度突出了传统RC架构丰富的储层状态。此外，评估了掩码对RC实现的影响，展示了该器件进行时空信号分析的能力。这一进展为实现节能且高性能的计算解决方案铺平了道路。
