An in-memory computing architecture based on a duplex two-dimensional material structure for in situ machine learning

Ning, Hongkai; Yu, Zhihao; Zhang, Qingtian; Wen, Hengdi; Gao, Bin; Mao, Yingchi; Li, Yuankun; Zhou, Ying; Zhou, Yue; Chen, Jiewei; Liu, Lei; Wang, Wenfeng; Li, Taotao; Li, Yating; Meng, Wanqing; Li, Weisheng; Liu, Xinyi; Qiu, Hao; Shi, Yi; Chai, Yang; Wu, Huaqiang; Wang, Xinran

doi:10.1038/s41565-023-01343-0

Cited by 39 publications

(19 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work is a big leap forward in overcoming the bottlenecks of modern nonvolatile memory. To further reduce program/erase times and voltages of ultrafast nonvolatile memory devices, HZO could be investigated as an alternative to PZT. , Our work has laid a solid foundation for high-performance memories based on emerging 2D materials.…”

Section: Discussionmentioning

confidence: 99%

Polarized Tunneling Transistor for Ultrafast Memory

Chen

Dun

et al. 2023

ACS Nano

View full text Add to dashboard Cite

In today's information age, high performance nonvolatile memory devices have become extremely important. Despite their potential, existing devices suffer from limitations, such as low operation speed, low memory capacity, short retention time, and a complex preparation process. To overcome these limitations, advanced memory designs are required to improve speed, memory capacity, and retention time and reduce the number of preparation steps. Here, we present a nonvolatile floating-gate-like memory device based on a transistor that uses the polarization effect of ferroelectric material PZT (Pb[Zr 0.2 Ti 0.8 ]O 3 ) for regulating tunneling electrons for charging and discharging the MoS 2 channel layer. The transistor is defined as a polarized tunneling transistor (PTT) and does not require a tunnel layer or a floating-gate layer. The PTT demonstrates an ultrafast programming/erasing speed of 25/20 ns and a response time of 120/105 ns, which is comparable to the ultrafast flash memories based on van der Waals heterostructures. Additionally, the PTT has a high extinction ratio of 10 4 , a long retention time of 10 years, and a simple fabrication process. Our research provides future guidelines for the development of the next generation of ultrafast nonvolatile memory devices.

show abstract

Section: Discussionmentioning

confidence: 99%

Polarized Tunneling Transistor for Ultrafast Memory

Chen

Dun

et al. 2023

ACS Nano

View full text Add to dashboard Cite

show abstract

“…The direct actuation of the mechanical platform using the artificial NMJ provides a wide range of neuromorphic sensing-to-action applications, including time-offlight ranging (38)(39)(40)(41)(42)(43), in-sensor/near-sensor computing (44)(45)(46)(47)(48), and human-computer interaction (49). In this study, we achieved a normalized output current of 200 mA/mm with the CIPS/GaN FeHEMT, which is notably greater than that of recently reported synaptic transistors (for more details, see table S1) (50)(51)(52)(53)(54)(55)(56)(57)(58)(59). Therethe CIPS/GaN FeHEMT is potentially deployable as an artificial NMJ in robotic systems to operate mechanical actuators that require a milliampere-scale driving current for macro-motion.…”

Section: Discussionmentioning

confidence: 76%

An artificial neuromuscular junction for enhanced reflexes and oculomotor dynamics based on a ferroelectric CuInP ₂ S ₆ /GaN HEMT

Park,

Yang,

Yeom

et al. 2023

Sci. Adv.

View full text Add to dashboard Cite

A neuromuscular junction (NMJ) is a particularized synapse that activates muscle fibers for macro-motions, requiring more energy than computation. Emulating the NMJ is thus challenging owing to the need for both synaptic plasticity and high driving power to trigger motions. Here, we present an artificial NMJ using CuInP 2 S 6 (CIPS) as a gate dielectric integrated with an AlGaN/GaN-based high-electron mobility transistor (HEMT). The ferroelectricity of the CIPS is coupled with the two-dimensional electron gas channel in the HEMT, providing a wide programmable current range of 6 picoampere per millimeter to 5 milliampere per millimeter. The large output current window of the CIPS/GaN ferroelectric HEMT (FeHEMT) allows for amplifier-less actuation, emulating the biological NMJ functions of actuation and synaptic plasticity. We also demonstrate the emulation of biological oculomotor dynamics, including in situ object tracking and enhanced stimulus responses, using the fabricated artificial NMJ. We believe that the CIPS/GaN FeHEMT offers a promising pathway for bioinspired robotics and neuromorphic vision.

show abstract

“…At the macroscopic scale, ferroelectric films can generate multiple polarized states through their multiple domains, resulting in the production of multiple-bit channel conductance states. As a result, FEFET can be used to mimic synaptic weight updating with multibit states. − …”

Section: Artificial Synapses and Neuronsmentioning

confidence: 99%

Artificial Neuron Devices

He,

Wang,

et al. 2023

Chem. Rev.

View full text Add to dashboard Cite

Efforts to design devices emulating complex cognitive abilities and response processes of biological systems have long been a coveted goal. Recent advancements in flexible electronics, mirroring human tissue’s mechanical properties, hold significant promise. Artificial neuron devices, hinging on flexible artificial synapses, bioinspired sensors, and actuators, are meticulously engineered to mimic the biological systems. However, this field is in its infancy, requiring substantial groundwork to achieve autonomous systems with intelligent feedback, adaptability, and tangible problem-solving capabilities. This review provides a comprehensive overview of recent advancements in artificial neuron devices. It starts with fundamental principles of artificial synaptic devices and explores artificial sensory systems, integrating artificial synapses and bioinspired sensors to replicate all five human senses. A systematic presentation of artificial nervous systems follows, designed to emulate fundamental human nervous system functions. The review also discusses potential applications and outlines existing challenges, offering insights into future prospects. We aim for this review to illuminate the burgeoning field of artificial neuron devices, inspiring further innovation in this captivating area of research.

show abstract

An in-memory computing architecture based on a duplex two-dimensional material structure for in situ machine learning

Cited by 39 publications

References 50 publications

Polarized Tunneling Transistor for Ultrafast Memory

Polarized Tunneling Transistor for Ultrafast Memory

An artificial neuromuscular junction for enhanced reflexes and oculomotor dynamics based on a ferroelectric CuInP ₂ S ₆ /GaN HEMT

Artificial Neuron Devices

Contact Info

Product

Resources

About

An in-memory computing architecture based on a duplex two-dimensional material structure for in situ machine learning

Cited by 39 publications

References 50 publications

Polarized Tunneling Transistor for Ultrafast Memory

Polarized Tunneling Transistor for Ultrafast Memory

An artificial neuromuscular junction for enhanced reflexes and oculomotor dynamics based on a ferroelectric CuInP 2 S 6 /GaN HEMT

Artificial Neuron Devices

Contact Info

Product

Resources

About

An artificial neuromuscular junction for enhanced reflexes and oculomotor dynamics based on a ferroelectric CuInP ₂ S ₆ /GaN HEMT