RBF Network Based on Improved Niche Hybrid Hierarchy Genetic Algorithm

2012 ◽  
Vol 433-440 ◽  
pp. 775-780
Author(s):  
Fang Wang ◽  
Jin Lan Yu ◽  
Pin Chang Zhu ◽  
Xi Feng Wei
Keyword(s):  
Genetic Algorithm ◽  
Time Series ◽  
Chaotic Time Series ◽  
Premature Convergence ◽  
Rbf Network ◽  
Operation Speed ◽  
System Resource ◽  
Simulation Results

The improved niche hybrid hierarchy genetic algorithm is presented to overcome the premature convergence which happens in genetic algorithm constructing RBF network. The niche with poor fitness of every individual is eliminated to save system resource and raise operation speed. The simulation results demonstrate the better predicted performance on the Mackey-Glass chaotic time series than other algorithms.

Evolutionary Computation Based Sizing Technique of Nulling Resistor Compensation Based CMOS Two-Stage Op-Amp Circuit

2017 ◽  
Vol 26 (04) ◽  
pp. 1740021 ◽  
Author(s):  
Bishnu Prasad De ◽  
Kanchan Baran Maji ◽  
Rajib Kar ◽  
Durbadal Mandal ◽  
Sakti Prasad Ghoshal
Keyword(s):  
Genetic Algorithm ◽  
Evolutionary Computation ◽  
Power Dissipation ◽  
Premature Convergence ◽  
Mos Transistors ◽  
Two Stage ◽  
Design Constraints ◽  
Op Amp ◽  
Simulation Results

This article explores the comparative optimizing efficiency between two PSO variants, namely, Craziness based PSO (CRPSO) and PSO with an Aging Leader and Challengers (ALC-PSO) for the design of nulling resistor compensation based CMOS two-stage op-amp circuit. The concept of PSO is simple and it replicates the nature of bird flocking. As compared with Genetic algorithm (GA), PSO deals with less mathematical operators. Premature convergence and stagnation problem are the two major limitations of PSO technique. CRPSO and ALC-PSO techniques individually have eliminated the disadvantages of the PSO technique. In this article, CRPSO and ALC-PSO are individually employed to optimize the sizes of the MOS transistors to reduce the overall area taken by the circuit while satisfying the design constraints. The results obtained individually from CRPSO and ALC-PSO techniques are validated in SPICE environment. SPICE based simulation results justify that ALC-PSO is much better technique than CRPSO and other formerly reported methods for the design of the afore mentioned circuit in terms of the MOS area, gain and power dissipation etc.

Multi Population Genetic Algorithm on Solving Multimodal Function

2013 ◽  
Vol 321-324 ◽  
pp. 2042-2046
Author(s):  
Hao Wen ◽  
Han Bin Chen
Keyword(s):  
Genetic Algorithm ◽  
Population Genetic ◽  
Extreme Value ◽  
Premature Convergence ◽  
Multimodal Function ◽  
Peak Function ◽  
Simulation Results ◽  
Population Genetic Algorithm ◽  
Standard Genetic Algorithm ◽  
The Way

This paper studies the way to solve extreme value of the nonlinear multi-peak function by using the multi-population genetic algorithm (MPGA). With the analysis of the advantages and defects of the standard genetic algorithm (SGA),the paper, This paper is use the population genetic algorithm to achieve the optimization and verification with Simulation for solving the extreme value of the nonlinear multi-peak function, which in order to achieve the solution with higher accuracy and higher efficiency. And make the analysis for the premature convergence that existed in SGA by comparing the standard genetic algorithm simulation results with that form the MPGA.

Algorithms Research in the Application of Lorenz Time Series Prediction

2011 ◽  
Vol 268-270 ◽  
pp. 1017-1020
Author(s):  
Man Xiang Miao ◽  
Yi Jin Gang
Keyword(s):  
Time Series ◽  
Statistical Learning ◽  
Time Series Prediction ◽  
Chaotic Time Series ◽  
Support Vector ◽  
Rbf Networks ◽  
Machine Learning Methods ◽  
Simulation Results ◽  
Time Series Simulation ◽  
Better Than

Prediction of Lorenz Chaotic Time Series is a vital problem in nonlinear dynamics .Support vector machine (SVM) is a kind of novel machine learning methods based on statistical learning theory, which have been provided an efficient algorithm thought in prediction of Chaotic Time Series. This paper combined SVM with neural network which based on the similarity of structure between SVM and RBF Networks, using SVM to obtain the centers of RBF Networks, then to predict the Lorenz Chaotic Time Series. Simulation results show that the effect is better than other methods.

scholarly journals Chaotic time series prediction with deep belief networks: an empirical evaluation

2020 ◽  
Vol 3 (SI1) ◽  
pp. SI102-SI112
Author(s):  
Duong Tuan Anh ◽  
Ta Ngoc Huy Nam
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Real World ◽  
Initial Conditions ◽  
Empirical Evaluation ◽  
Chaotic Time Series ◽  
Deep Belief Network ◽  
Support Vector ◽  
Belief Network ◽  
Rbf Network

Chaotic time series are widespread in several real world areas such as finance, environment, meteorology, traffic flow, weather. A chaotic time series is considered as generated from the deterministic dynamics of a nonlinear system. The chaotic system is sensitive to initial conditions; points that are arbitrarily close initially become exponentially further apart with progressing time. Therefore, it is challenging to make accurate prediction in chaotic time series. The prediction using conventional statistical techniques, k-nearest-nearest neighbors algorithm, Multi-Layer-Perceptron (MPL) neural networks, Recurrent Neural Networks, Radial-Basis-Function (RBF) Networks and Support Vector Machines, do not give reliable prediction results for chaotic time series. In this paper, we investigate the use of a deep learning method, Deep Belief Network (DBN), combined with chaos theory to forecast chaotic time series. DBN should be used to forecast chaotic time series. First, the chaotic time series are analyzed by calculating the largest Lyapunov exponent, reconstructing the time series by phase-space reconstruction and determining the best embedding dimension and the best delay time. When the forecasting model is constructed, the deep belief network is used to feature learning and the neural network is used for prediction. We also compare the DBN –based method to RBF network-based method, which is the state-of-the-art method for forecasting chaotic time series. The predictive performance of the two models is examined using mean absolute error (MAE), mean squared error (MSE) and mean absolute percentage error (MAPE). Experimental results on several synthetic and real world chaotic datasets revealed that the DBN model is applicable to the prediction of chaotic time series since it achieves better performance than RBF network.

Sign in / Sign up

Export Citation Format

Share Document