Neuromorphic Artificial Vision Systems Based on Reconfigurable Ion‐Modulated Memtransistors

Zhen, Yang; Zhang, Teng; Liu, Keqin; Dang, Bingjie; Xu, Lei; Yang, Yuchao; Huang, Ru

doi:10.1002/aisy.202300026

Cited by 5 publications

(1 citation statement)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this regard, neuromorphic devices, such as memristors, charge-based field effect transistors, ferroelectric field effect transistors, and electric double layer (EDL) transistors, have been extensively researched. [6][7][8][9][10][11] In particular, EDL-based transistors are considered suitable for mimicking synaptic plasticity owing to the large reversible capacitance produced by ions in electrolytes. [12,13] In addition, they have excellent short-term plasticity (STP) and long-term plasticity (LTP) characteristics at low voltages and have the ability to adjust recovery time depending on the type of ions.…”

Section: Introductionmentioning

confidence: 99%

Realization of High Mobility Synaptic Transistor through Control of Cross‐Linking Agent in a Polymer Dielectric Layer for Emerging Electric Double Layer

Kim,

Yun

et al. 2024

Adv Materials Technologies

View full text Add to dashboard Cite

The development of high‐mobility neuromorphic transistors is essential to increase signal transmission speed and solve the von Neumann bottleneck issue. Herein, this work proposes cross‐linked Poly(4‐vinylphenol) (c‐PVP) as a dielectric layer to form an electric double layer (EDL), which plays a key role in synaptic transistors, and measure the ratio‐dependent characteristics of cross‐linking agents in c‐PVP. When the ratio of PVP to poly(melamine‐co‐formaldehyde) methylated (PMF) is 10:1, neuromorphic transistors is found to show the best performance with a memory window of 2.2 V and a mobility of 93.4 cm2 V−1 s−1. Fourier‐transform infrared spectroscopy (FT‐IR) results show that the reduction in the concentration of the cross‐linking agent generates more hydroxyl groups within the c‐PVP film. The 10:1 c‐PVP‐based synaptic device has an ultra‐low energy consumption of 15.8 pJ for a single pulse and a maximum paired pulse facilitation (PPF) index value of 291%. Additionally, synaptic characteristics, such as pulse duration time dependent plasticity, pulse intensity dependent plasticity, pulse rate dependent plasticity (SRDP), high band filtering, and short‐term memory (STM) conversion to long‐term memory (LTM), are also described and discussed. These results suggest that c‐PVP/ZnON‐based neuromorphic devices can be promising artificial synapses for memory and learning capabilities.

show abstract

Section: Introductionmentioning

confidence: 99%

Realization of High Mobility Synaptic Transistor through Control of Cross‐Linking Agent in a Polymer Dielectric Layer for Emerging Electric Double Layer

Kim,

Yun

et al. 2024

Adv Materials Technologies

View full text Add to dashboard Cite

show abstract

Fully Hardware Memristive Neuromorphic Computing Enabled by the Integration of Trainable Dendritic Neurons and High‐Density RRAM Chip

Yang,

Yue,

Liu

et al. 2024

Adv Funct Materials

View full text Add to dashboard Cite

Computing‐in‐memory (CIM) architecture inspired by the hierarchy of human brain is proposed to resolve the von Neumann bottleneck and boost acceleration of artificial intelligence. Whereas remarkable progress has been achieved for CIM, making further improvements in CIM performance is becoming increasingly challenging, which is mainly caused by the disparity between rapid evolution of synaptic arrays and relatively slow progress in building efficient neuronal devices. Specifically, dedicated efforts are required toward developments of more advanced activation units in terms of both optimized algorithms and innovative hardware implementations. Here a novel bio‐inspired dendrite function‐like neuron based on negative‐differential‐resistance (NDR) behavior is reported and experimentally demonstrates this design as a more efficient neuron. By integrating electrochemical random‐access memory (ECRAM) with ionic regulation, the tunable NDR neuron can be trained to enhance neural network performances. Furthermore, based on a high‐density RRAM chip, fully hardware implementation of CIM is experimentally demonstrated by integrating NDR neuron devices with only a 1.03% accuracy loss. This work provides 516 × and 1.3 × 105 × improvements on LAE (Latency‐Area‐Energy) property, compared to the digital and analog CMOS activation circuits, respectively. With device‐algorithm co‐optimization, this work proposes a compact and energy‐efficient solution that pushes CIM‐based neuromorphic computing into a new paradigm.

show abstract

A flexible thermal-coupled InGaZnO adaptive synapse

Xu,

Long,

et al. 2024

Applied Physics Letters

View full text Add to dashboard Cite

The development of neuromorphic sensory systems necessitates synaptic devices with adaptivity to a wide range of stimuli. Furthermore, the introduction of multimodal adaptivity is highly favorable, which holds immense potential for improving the processing capability of the neuromorphic system under complex environments. In this work, we report a thermal-coupled adaptive synapse (TCAS) by integrating an IGZO-based synaptic transistor with a laser-induced graphene micro-heater. This synapse enables active modulation of nonlinear short-term plasticity gains through temperature and voltage co-mediated ion/electron coupling, which facilitates the high adaptivity for image denoising. The images with multilevel signals can be effectively denoised with an average reduction of ∼84.0% in the Euclidean distance in comparison with the noisy images. This outcome indicates the effectiveness of TCASs in complex sensory processing and offers a promising solution for the development of neuromorphic systems with multimodal adaptability.

show abstract

Neuromorphic Artificial Vision Systems Based on Reconfigurable Ion‐Modulated Memtransistors

Cited by 5 publications

References 37 publications

Realization of High Mobility Synaptic Transistor through Control of Cross‐Linking Agent in a Polymer Dielectric Layer for Emerging Electric Double Layer

Realization of High Mobility Synaptic Transistor through Control of Cross‐Linking Agent in a Polymer Dielectric Layer for Emerging Electric Double Layer

Fully Hardware Memristive Neuromorphic Computing Enabled by the Integration of Trainable Dendritic Neurons and High‐Density RRAM Chip

A flexible thermal-coupled InGaZnO adaptive synapse

Contact Info

Product

Resources

About