Related references
Note: Only part of the references are listed.
Article
Physics, Applied
Yi Huang et al.
Summary: Progress in AI hardware and algorithms has led to the development of large machine learning models and various applications in our daily lives. However, current AI, mainly artificial neural networks, still falls short of human brains due to high energy consumption and the inability to generalize knowledge and adapt to changes. Neuromorphic systems built with emerging devices like memristors offer a promising solution to these challenges. This Perspective provides an overview of the progress and challenges in transitioning from memristor devices to neuromorphic systems, and suggests potential directions for implementation based on memristive devices.
APPLIED PHYSICS LETTERS
(2023)
Article
Physics, Applied
Yang Yang et al.
Summary: Physical reservoir computing (PRC) is a potential low-cost temporal processing platform, utilizing the nonlinear and volatile dynamics of materials. An electric-double-layer (EDL) formed at the interface between a semiconductor and an electrolyte is used to build a high energy-efficiency PRC. In this study, EDL coupled indium-gallium-zinc-oxide (IGZO) artificial synapses are used to implement reservoir computing (RC), which exhibits nonlinearity, fade memory properties, and low power consumption, making it suitable for high energy-efficient RC systems. The results show the potential of using EDL coupling for a lightweight hardware physical reservoir that can underlie a next-generation machine learning platform.
APPLIED PHYSICS LETTERS
(2023)
Article
Physics, Applied
Aleksandra A. Koroleva et al.
Summary: The research investigates the neuromorphic capabilities of a self-aligned complementary metal-oxide-semiconductor compatible W/WOx/HfO2/Ru device in a 3D vertical memristive structure. The device demonstrates nonfilamentary forming-free multilevel resistive switching and gradual resistance change. It can be integrated into architectures that require short-term memory characteristics, such as reservoir computing systems. The device's ability to rely on the temporal sequence of the stream has been tested with the digit recognition task. With thermal oxidization and atomic layer deposition methods used for layer formation, the device is suitable for high-density systems with high connectivity in neural networks.
APPLIED PHYSICS LETTERS
(2023)
Article
Nanoscience & Nanotechnology
Huiwu Mao et al.
Summary: This research introduces a stacked memristor based on IGZO as a spiking stochastic neuron, showing tunable firing probability and demonstrating eliminated switching variation and small relative deviation in stacked configuration compared to a single memristor.
ADVANCED ELECTRONIC MATERIALS
(2022)
Article
Chemistry, Multidisciplinary
Rui Wang et al.
Summary: A study reports on the use of indium-galim-zinc-oxide thin film phototransistors for in-sensor compression and computing. The results show that the system achieves efficient data processing while maintaining high recognition accuracy.
Article
Multidisciplinary Sciences
Seokhyeong Lee et al.
Summary: The authors report the realization of a programmable image sensor based on black phosphorus that can perform multispectral imaging and analog in-memory computing in the near- to mid-infrared range. The sensor enables in-sensor computing, reducing communication latency and power consumption. This multifunctional infrared image sensor has great potential for distributed and remote multispectral sensing applications.
NATURE COMMUNICATIONS
(2022)
Article
Multidisciplinary Sciences
Weier Wan et al.
Summary: This study presents NeuRRAM, a RRAM-based CIM chip that offers versatility, high energy efficiency, and accuracy. By co-optimizing algorithms, architecture, circuits, and devices, the chip can be reconfigured for different model architectures and provides twice the energy efficiency of previous state-of-the-art RRAM-CIM chips across various computational bit-precisions. It achieves inference accuracy comparable to software models quantized to four-bit weights across various AI tasks.
Article
Multidisciplinary Sciences
Akhil Dodda et al.
Summary: The authors propose a solution for IoT edge sensors based on two-dimensional (2D) memtransistors, integrating sensing, storage, and encryption functionalities. The sensors offer low power consumption and near-sensor security.
NATURE COMMUNICATIONS
(2022)
Article
Physics, Applied
Md. Fahim F. Chowdhury et al.
Summary: We demonstrate that a nanomagnet array excited by surface acoustic waves can serve as a reservoir with high memory effect and computing capability. The nanomagnet array has lower energy dissipation and is able to accurately predict time series with scalability and flexibility.
APPLIED PHYSICS LETTERS
(2022)
Article
Physics, Applied
Jingjuan Wang et al.
Summary: Electronic synaptic devices with photoelectric sensing function are important in the development of neuromorphic computing system. This study presents a photoelectrical synaptic system based on high-quality epitaxial films and demonstrates its potential in simulating synaptic behavior and achieving visual perception functions.
APPLIED PHYSICS LETTERS
(2022)
Article
Multidisciplinary Sciences
Yanan Zhong et al.
Summary: Zhong et al. developed a parallel memristor-based reservoir computing system capable of tuning critical parameters, achieving a classification accuracy of 99.6% in spoken-digit recognition and a time-series prediction error of 0.046 in the Henon map.
NATURE COMMUNICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Cheng-Xin Xue et al.
Summary: The development of small, energy-efficient artificial intelligence edge devices has been limited by data transfer requirements between the processor and memory in traditional computing architectures. Non-volatile compute-in-memory (nvCIM) architectures show potential to overcome these limitations, but challenges still remain in developing configurations for high-bit-precision dot-product operations.
NATURE ELECTRONICS
(2021)
Article
Multidisciplinary Sciences
Linfeng Sun et al.
Summary: This study presents a sensor reservoir computing approach for language learning, utilizing two-dimensional memristors to achieve high dimensionality, nonlinearity, and fading memory. With an accuracy of 91% in classifying short sentences, it offers a low training cost and real-time solution for processing temporal and sequential signals in machine learning applications at the edge.
Article
Multidisciplinary Sciences
Daniel J. Gauthier et al.
Summary: Reservoir computers are artificial neural networks that can be trained on small data sets with large random matrices and numerous metaparameters. Nonlinear vector autoregression is a superior machine learning algorithm compared to reservoir computing, requiring fewer training data sets and training time.
NATURE COMMUNICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Wen-Qian Pan et al.
IEEE TRANSACTIONS ON ELECTRON DEVICES
(2020)
Article
Multidisciplinary Sciences
Peng Yao et al.
Article
Multidisciplinary Sciences
Tianda Fu et al.
NATURE COMMUNICATIONS
(2020)
Review
Computer Science, Artificial Intelligence
Gouhei Tanaka et al.
Article
Engineering, Electrical & Electronic
Wei Wang et al.
IEEE ELECTRON DEVICE LETTERS
(2019)
Article
Automation & Control Systems
Rivu Midya et al.
ADVANCED INTELLIGENT SYSTEMS
(2019)
Article
Engineering, Electrical & Electronic
John Moon et al.
NATURE ELECTRONICS
(2019)
Article
Materials Science, Multidisciplinary
Sanghyeon Choi et al.
NPG ASIA MATERIALS
(2018)
Article
Multidisciplinary Sciences
Chao Du et al.
NATURE COMMUNICATIONS
(2017)
Article
Multidisciplinary Sciences
Peng Yao et al.
NATURE COMMUNICATIONS
(2017)
Article
Computer Science, Artificial Intelligence
Ali Rodan et al.
IEEE TRANSACTIONS ON NEURAL NETWORKS
(2011)
Article
Multidisciplinary Sciences
L. Appeltant et al.
NATURE COMMUNICATIONS
(2011)
Article
Chemistry, Multidisciplinary
Sung Hyun Jo et al.
Article
Computer Science, Information Systems
Mantas Lukosevicius et al.
COMPUTER SCIENCE REVIEW
(2009)
Article
Multidisciplinary Sciences
H Jaeger et al.
