338,889 Hits in 3.2 sec

The time dimension of neural network models

Richard Rohwer
1994 ACM SIGART Bulletin  
This review attempts to provide an insightful perspective on the role of time within neural network models and the use of neural networks for problems involving time.  ...  The most commonly used neural network models are de ned and explained giving mention to important technical issues but avoiding great detail.  ...  Basic Neural Network Models Neural network models specify rules for changing the output values of model neurons, or nodes, with time.  ... 
doi:10.1145/181911.181917 fatcat:qk5nunmmrzcohncrnm4k7wd3uq

Prediction Model of Weekly Retail Price for Eggs Based on Chaotic Neural Network

Zhe-min LI, Li-guo CUI, Shi-wei XU, Ling-yun WENG, Xiao-xia DONG, Gan-qiong LI, Hai-peng YU
2013 Journal of Integrative Agriculture  
Based on the weekly retail prices of eggs from Jan. 2008 to Dec. 2012, this paper establishes a short-term prediction model of weekly retail prices of eggs based on chaotic neural network.  ...  In the process of determining the structure of the chaotic neural network, the number of input layer nodes of the network is calculated by reconstructing phase space and computing its saturated embedding  ...  Fig. 7 . 7 Actual results and simulation results of the chaotic neural network The prediction results comparison between the chaotic neural network model and the traditional time series model (ARMA model  ... 
doi:10.1016/s2095-3119(13)60610-3 fatcat:ckivaw2mb5dwdo3maguaq3qkpu

Real Estate Price Prediction Model Based on Dynamic Neural Network

2016 Revista Técnica de la Facultad de Ingeniería Universidad del Zulia  
This paper presents a kind of real estate price prediction model that is based on the theory of the dynamic neural network.  ...  This algorithm firstly constructs the multidimensional time series of BP network topology; it then respectively constructs the residential real estate price time series of prediction model for different  ...  prediction model based on BP neural network time series toolbox.  ... 
doi:10.21311/ fatcat:3m2mccz6fnae5ide23hqfta2ki

Effect of various dimension convolutional layer filters on traffic sign classification accuracy

V.N. Sichkar, S.A. Kolyubin
2019 Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki  
Every model of convolutional neural network has the same architecture but different dimension of filters for convolutional layer.  ...  Abstract The paper presents the study of an effective classification method for traffic signs on the basis of a convolutional neural network with various dimension filters.  ...  The convolutional neural network training takes place with batches of 50 examples at the same time.  ... 
doi:10.17586/2226-1494-2019-19-3-546-552 fatcat:iw3u6r6e6jgmrafvog7s24d4aq

Combined Prediction of Wind Power with Chaotic Time Series Analysis

Wang Qiang, Yang Yang
2014 Open Automation and Control Systems Journal  
The results show that the proposed model is more effective than single embedding dimension model and linear weighted combination model, and the prediction error of neural network combination is less than  ...  A combined model for wind power forecasting is presented to decrease the influence of reconstructed parameters by chaotic time series analysis and the neural networks (NNs) in this work.  ...  phase space is better than neural network model with single phase space reconstruction; 4) embedding dimension, namely, the input number for neural network has great impact on the prediction accuracy;  ... 
doi:10.2174/1874444301406010117 fatcat:f47rxsksd5eabgbgucjynb4nfe

Neural network method for determining embedding dimension of a time series

A. Maus, J.C. Sprott
2011 Communications in nonlinear science & numerical simulation  
the sensitivity of the output of the network to each time lag averaged over the data set.  ...  A method is described for determining the optimal short term prediction time-delay embedding dimension for a scalar time series by training an artificial neural network on the data and then determining  ...  It is remarkable that with only four neurons, the neural network is able to accurately model two co-mingled two-dimensional nonlinear maps.  ... 
doi:10.1016/j.cnsns.2010.10.030 fatcat:iombpyrc6fdtbko72nutpyb3ri

ASAT: Adaptively Scaled Adversarial Training in Time Series [article]

Zhiyuan Zhang, Wei Li, Ruihan Bao, Keiko Harimoto, Yunfang Wu, Xu Sun
2021 arXiv   pre-print
Besides enhancing neural networks, we also propose the dimension-wise adversarial sensitivity indicator to probe the sensitivities and importance of input dimensions.  ...  Besides the security concerns of potential adversarial examples, adversarial training can also improve the performance of the neural networks, train robust neural networks, and provide interpretability  ...  The experiment data is provided by Mizuho Securities and Reuters.  ... 
arXiv:2108.08976v1 fatcat:pvdid3up2fa3hltqdnd6bfu34u

Development of an artificial neural network to predict lead frame dimensions in an etching process

Tzu-Chiang Liu, Rong-Kwei Li, Meng-Chi Chen
2005 The International Journal of Advanced Manufacturing Technology  
This study presents the development of an artificial neural network (ANN) model that can be applied to construct the predicting model.  ...  The predictive model can estimate the dimensions of the pilot hole and thus determine the process parameters needed to improve lead frame quality in the etching process.  ...  The node is the predicted value of the t + 1th period. Figure 6 is the feedforward neural network model for a time series. Neural networks have been widely used for forecasting problems [10] .  ... 
doi:10.1007/s00170-004-2310-5 fatcat:d4hn3yagqzdcbccv26da7c3lcm

Comparison between Neural Network and Adaptive Neuro-Fuzzy Inference System for Forecasting Chaotic Traffic Volumes

Jiin-Po Yeh, Yu-Chen Chang
2012 Journal of Intelligent Learning Systems and Applications  
The architecture of the neural network consists of the input vector, one hidden layer and output layer.  ...  The input variables and target of the adaptive neuro-fuzzy inference system are the same as those of the neural network.  ...  of next state as the target of the neural network.  ... 
doi:10.4236/jilsa.2012.44025 fatcat:f6epuvv6tbdi7ofaifx6ejwnpm

Evaluating Lyapunov exponent spectra with neural networks

A. Maus, J.C. Sprott
2013 Chaos, Solitons & Fractals  
The method uses a distribution of calculated exponent values produced by modeling a single time series many times or multiple instances of a time series.  ...  An additional merit of global modeling is its ability to estimate the dynamical and geometrical properties of the original system such as the attractor dimension, entropy, and lag space, although consideration  ...  Neural networks build a global model of the data, but there is a trade-off between the amount of computation required and accuracy of the model.  ... 
doi:10.1016/j.chaos.2013.03.001 fatcat:6k2l4t3f3jbzrj42y3ufeuf4t4

Solar Radiation Prediction Based on Phase Space Reconstruction of Wavelet Neural Network

Jianping Wang, Yunlin Xie, Chenghui Zhu, Xiaobing Xu a
2011 Procedia Engineering  
It's hard to model with a single method. A wavelet neural network model was set in this paper.  ...  The nonlinear process of solar radiation was forecasted by neural network and the non-stationary process of solar radiation was decomposed into quasi-stationary at different frequency scales by multi-scale  ...  Acknowledgements The authors wish to thank the data providers that have led to the analysis being possible in the presentation of the paper.  ... 
doi:10.1016/j.proeng.2011.08.864 fatcat:x7pz2mnhrvfgpasxvyrbtw7soe

Sunspot Forecasting by Using Chaotic Time-series Analysis and NARX Network

Chuanjin Jiang, Fugen Song
2011 Journal of Computers  
Take annual active times of sunspot as an example, after verifying the chaos of sunspot time-series and calculating the series' embedding dimension and delay, we establish sunspot prediction model with  ...  Chaotic time-series is a dynamic nonlinear system whose features can not be fully reflected by Linear Regression Model or Static Neural Network.  ...  ACKNOWLEDGMENT The authors wish to thank Wenguan Song and Zhang Jin.  ... 
doi:10.4304/jcp.6.7.1424-1429 fatcat:tlxz2k7ghnchfatk7zjjfq3qvy

Reinforcement Learning for Improving the Accuracy of PM2.5 Pollution Forecast Under the Neural Network Framework

Shih Wei Liao, Shuo Wen Chang, Long-Tin Li
2019 IEEE Access  
Second, we select the input with different input dimensions and values of time delay, calculate the best strategy, and evaluate the computational complexity of our RL algorithm.  ...  Some predecessors have experimented with artificial neural networks (NNs), combining linear autoregressive integrated moving average (ARIMA) models with nonlinear NN models.  ...  The focus of our work is to use its characteristics to select the input dimension of the neural network model and the time delay between inputs.  ... 
doi:10.1109/access.2019.2932413 fatcat:hxyumtisungkjhpkwlytaqy5tu

Incomplete Phase Space Reconstruction Method Based on Subspace Adaptive Evolution Approximation

Tai-fu Li, Wei Jia, Wei Zhou, Ji-ke Ge, Yu-cheng Liu, Li-zhong Yao
2013 Journal of Applied Mathematics  
The subspace approximation of neural network based on the nonlinear extended Kalman filtering (EKF) is a dynamic evolution approximation from one neighborhood to another.  ...  The common static neural network approximation is suitable for a trained neighborhood, but it cannot ensure its generalization performance in other untrained neighborhood.  ...  Acknowledgments This work was supported by the National Science Foundation of China (no. 51075418), the National Science Foundation of China (no. 61174015), Chongqing CMEC Foundations of China (no.  ... 
doi:10.1155/2013/983051 fatcat:qshsjxf2svhldh3z6ofancx6cy

Research on Neural Machine Translation Model

Mengyao Chen, Yong Li, Runqi Li
2019 Journal of Physics, Conference Series  
mode of convolutional neural network or circular neural network and only uses the self-attention mechanism.  ...  Convolution neural network has replaced the divine circulation neural network due to its parallel computation of convolution.  ...  Acknowledgments we express our sincere gratitude to Teacher Li Yong for his help in the process of writing the thesis.  ... 
doi:10.1088/1742-6596/1237/5/052020 fatcat:nghf3oryznatboysa2t4xswlmu
« Previous Showing results 1 — 15 out of 338,889 results