An echo state network architecture based on quantum logic gate and its optimization

Liu, Junxiu; Sun, Tiening; Luo, Yuling; Yang, Su; Cao, Yi; Zhai, Jia

doi:10.1016/j.neucom.2019.09.002

Cited by 27 publications

(14 citation statements)

References 31 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The SNNcube is a scalable module. The leaky integrate-and-fire (LIF) neurons are used to implement the SNNCube [33], [34]. The parameters of n x , n y and n z are used to control the size of SNNcube, which represent the number of neurons along the x, y and z axes, respectively.…”

Section: Emotion Classification Using the Snnmentioning

confidence: 99%

EEG-Based Emotion Classification Using Spiking Neural Networks

et al. 2020

Self Cite

View full text Add to dashboard Cite

A novel method of using the spiking neural networks (SNNs) and the electroencephalograph (EEG) processing techniques to recognize emotion states is proposed in this paper. Three algorithms including discrete wavelet transform (DWT), variance and fast Fourier transform (FFT) are employed to extract the EEG signals, which are further taken by the SNN for the emotion classification. Two datasets, i.e., DEAP and SEED, are used to validate the proposed method. For the former dataset, the emotional states include arousal, valence, dominance and liking where each state is denoted as either high or low status. For the latter dataset, the emotional states are divided into three categories (negative, positive and neutral). Experimental results show that by using the variance data processing technique and SNN, the emotion states of arousal, valence, dominance and liking can be classified with accuracies of 74%, 78%, 80% and 86.27% for the DEAP dataset, and an overall accuracy is 96.67% for the SEED dataset, which outperform the FFT and DWT processing methods. In the meantime, this work achieves a better emotion classification performance than the benchmarking approaches, and also demonstrates the advantages of using SNN for the emotion state classifications.

show abstract

Section: Emotion Classification Using the Snnmentioning

confidence: 99%

EEG-Based Emotion Classification Using Spiking Neural Networks

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…After the reservoir state and the output mode of the ESN are determined, can be determined according to the target output, so that the error between ( ) and the target output is as small as possible. This is a simple linear regression problem and many methods can be used to calculate it [24]. In addition, ESN has a good advantage for the processing of one-dimensional time series.…”

Section: Esnmentioning

confidence: 99%

Echo state network optimization using binary grey wolf algorithm

et al. 2020

Self Cite

View full text Add to dashboard Cite

The echo state network (ESN) is a powerful recurrent neural network for time series modelling. ESN inherits the simplified structure and relatively straightforward training process of conventional neural networks, and shows strong computational capabilities to solve nonlinear problems. It is able to map low-dimensional input signals to high-dimensional space for information extraction, but it is found that not every dimension of the reservoir output directly contributes to the model generalization. This work aims to improve the generalization capabilities of the ESN model by reducing the redundant reservoir output features. A novel hybrid model, namely binary grey wolf echo state network (BGWO-ESN), is proposed which optimises the ESN output connection by the feature selection scheme. Specially, the feature selection scheme of BGWO is developed to improve the ESN output connection structure. The proposed method is evaluated using synthetic and financial data sets. Experimental results demonstrate that the proposed BGWO-ESN model is more effective than other benchmarks, and obtains the lowest generalization error.

show abstract

“…Появление гибридных нейросетевых решений вызвано желанием преодолеть ограничения отдельных архитектур [22][23][24][25]. В работе [22] предложен подход, основанный на комбинировании долгой краткосрочной памяти и сверточных нейронных сетей.…”

Section: научные статьиunclassified

“…В [23] используются глубокие сети доверия. Гибрид нейронных сетей состояний «эхо» и квантовых сетей предложен в [24]. В [25] рассмотрен подход, основанный на объединении НС прямого распространения и нечетких систем.…”

Section: научные статьиunclassified

Neural network event forecasting for robots with continuous training

Osipov¹,

Miloserdov

2020

ICS

View full text Add to dashboard Cite

Introduction: High hopes for a significant expansion of human capabilities in various fields of activity are pinned on the creation and use of highly intelligent robots. To achieve this level of robot intelligence, it is necessary to successfully solve the problems of predicting the external environment and the state of the robots themselves. Solutions based on recurrent neural networks with controlled elements are promising neural network forecasting systems. Purpose: Search for appropriate neural network structures for predicting events. Development of approaches to controlling the associative call of information from a neural network memory. Methods: Computer simulation of recurrent neural networks with controlled elements and various structures of layers. Results: An improved method of neural network event forecasting with continuous robot training has been developed. This method allows you to predict events on either long or short samples of time series. In order to improve the forecasting accuracy, new rules have been proposed for controlling the associative call of information from the neural network memory. A software system has been developed which implements the proposed method and supports the emulation of neural networks with various layer structures. The possibilities of recurrent neural networks with linear or spiral layer structures are analyzed using the example of urban traffic flow forecasting. The gain of the proposed method in comparison with the ARIMA model for the MAPE indicator is from 4.1 to 7.4%. Among the studied neural network structures, the spiral structures have shown the highest accuracy, and linear structures have shown the lowest accuracy. Practical relevance: The results of the study can be used to improve the accuracy of event forecasting for intelligent robots.

show abstract

An echo state network architecture based on quantum logic gate and its optimization

Abstract: Ti t l e An e c h o s t a t e n e t w o r k a r c hi t e c t u r e b a s e d o n q u a n t u m lo gic g a t e a n d its o p ti miz a tio n

Cited by 27 publications

References 31 publications

EEG-Based Emotion Classification Using Spiking Neural Networks

EEG-Based Emotion Classification Using Spiking Neural Networks

Echo state network optimization using binary grey wolf algorithm

Neural network event forecasting for robots with continuous training

Contact Info

Product

Resources

About