Memristor-Based LSTM Network for Text Classification

Dou, Gang; Zhao, Kaixuan; Guo, Mei; Mou, Jun

doi:10.1142/s0218348x23400406

Cited by 42 publications

(14 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By using the LSTM network, the CC value in the NARMA2 task comparable to that obtained by using IL-PRD was obtained (Figure S19). However, in order to implement the LSTM unit by physical devices, a large number of the resistance switching devices (memristors) are required. , Therefore, the introduction of IL-PRD having hysteretic and nonlinear transformation characteristics leads to the size reduction of the PRC system, which is favorable for the application in the field of edge computing.…”

Section: Resultsmentioning

confidence: 99%

Characterization of Information-Transmitting Materials Produced in Ionic Liquid-based Neuromorphic Electrochemical Devices for Physical Reservoir Computing

Sato,

Shima,

Matsuo

et al. 2023

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Device implementation of reservoir computing, which is expected to enable high-performance data processing in simple neural networks at a low computational cost, is an important technology to accelerate the use of artificial intelligence in the real-world edge computing domain. Here, we propose an ionic liquid-based physical reservoir device (IL-PRD), in which copper cations dissolved in an IL induce diverse electrochemical current responses. The origin of the electrochemical current from the IL-PRD was investigated spectroscopically in detail. After operating the device under various operating conditions, X-ray photoelectron spectroscopy of the IL-PRD revealed that electrochemical reactions involving Cu, Cu 2 O, Cu(OH) 2 , CuS x , and H 2 O occur at the Pt electrode/IL interface. These products are considered information transmission materials in IL-PRD similar to neurotransmitters in biological neurons. By introducing the Faradaic current components due to the electrochemical reactions of these materials into the output signal of IL-PRD, we succeeded in improving the time-series data processing performance of the nonlinear autoregressive moving average task. In addition, the information processing efficiency in machine learning to classify electrocardiogram signal waveforms was successfully improved by using the output current from IL-PRD. Optimizing the electrochemical reaction products of IL-PRD is expected to advance data processing technology in society.

show abstract

Section: Resultsmentioning

confidence: 99%

Characterization of Information-Transmitting Materials Produced in Ionic Liquid-based Neuromorphic Electrochemical Devices for Physical Reservoir Computing

Sato,

Shima,

Matsuo

et al. 2023

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

show abstract

“…The data-driven model based SOC estimation methods.-The data-driven model is based on the black box principle and big data, 15,20 without considering the complex reaction process inside the battery, it is based on a large number of lithium-ion battery test data to establish the relationship between input and output. We only need to design reasonable training methods and rely on the collected experimental data to train and test the model.…”

Section: Soc Estimation Methodsmentioning

confidence: 99%

A Hierarchical Identification Method for Lithium-Ion Battery SOC Based on the Hammerstein Model

Wang¹,

Ding²,

Wang³

2023

J. Electrochem. Soc.

View full text Add to dashboard Cite

The two-input one-output Hammerstein model consists of two parallel nonlinear static blocks followed by a linear dynamic part. By using Hammerstein structure to map the relationship between a battery state of charge (SOC) and its terminal voltage/current, a hierarchical stochastic gradient algorithm is studied to estimate parameters of Hammerstein SOC model, so as to predict battery SOC. First, the Hammerstein model is transformed into a bilinear parameter system with the least number of required parameters. Then, a hierarchical stochastic gradient algorithm with a forgetting factor is used to update the two sets of model parameters of the bilinear parameter system, so as to realize SOC estimation. Furthermore, the experiment platform of lithium-ion battery was built and the data of the urban dynamometer driving schedule profile and the Los Angeles 92 profile were collected. Finally, the MATLAB simulation results show that the proposed parameter optimized method based Hammerstein model has the advantages of fast convergence speed and high SOC estimation accuracy.

show abstract

“…For a fixed set of parameters, the system trajectory may move to different stable states for different initial states [24][25][26][27]. Generally, the attractor coexistence phenomenon is easily found in memory element circuits and trigonometric systems [28][29][30][31][32]. In particular, the systems with an infinitely large number of coexisting attractors are known as extreme multistability [33][34][35].…”

Section: Introductionmentioning

confidence: 99%

A new fractional-order complex chaotic system with extreme multistability and its implementation

Ren¹,

Li²,

Banerjee³

et al. 2023

Phys. Scr.

Self Cite

View full text Add to dashboard Cite

In this paper, a new fractional-order complex chaotic system (FOCCS) is proposed and studied. Firstly, the dissipativity and stability are discussed. Secondly, the dynamical characteristics of the system with parameters and order changes are analyzed by using phase diagrams, Lyapunov exponent (LEs) and bifurcation diagrams, respectively. In addition, the dynamical behavior is discussed for q of integer and fractional orders. In particular, the attractor coexistence is found, such as the coexistence of chaotic attractor and chaotic attractor, and chaotic attractor and periodic attractor. Interestingly, the multiple attractors coexistence is found by changing the initial conditions with ﬁxed parameters. Finally, it is implemented on the analog circuit and DSP platform. The study provide a reference for the research and application of chaos.

show abstract

Memristor-Based LSTM Network for Text Classification

Cited by 42 publications

References 35 publications

Characterization of Information-Transmitting Materials Produced in Ionic Liquid-based Neuromorphic Electrochemical Devices for Physical Reservoir Computing

Characterization of Information-Transmitting Materials Produced in Ionic Liquid-based Neuromorphic Electrochemical Devices for Physical Reservoir Computing

A Hierarchical Identification Method for Lithium-Ion Battery SOC Based on the Hammerstein Model

A new fractional-order complex chaotic system with extreme multistability and its implementation

Contact Info

Product

Resources

About