Hierarchical genetic algorithm based neural network design

Published 7 November 2002

Gary G. Yen, Haiming Lu

DOI10.1109/ecnn.2000.886232

Citations37

TL;DR

The proposed Hierarchical genetic algorithm based design procedure for multi-layer feedforward neural network proves to be competitive or even superior to the traditional back-propagation network in Mackey-Glass chaotic time series prediction.

Abstract

In this paper, we propose a novel genetic algorithm based design procedure for multi-layer feedforward neural network. Hierarchical genetic algorithm is used to evolve both neural network topology and parameters. Compared with traditional genetic algorithm based designs for neural network, the proposed hierarchical approach addressed several deficiencies highlighted in literature. A multi-objective function is used herein to optimize the performance and topology of the evolved neural network. Two benchmark problems are successfully verified and the proposed algorithm proves to be competitive or even superior to the traditional back-propagation network in Mackey-Glass chaotic time series prediction.

Keywords

Computer ScienceEngineering

Neural NetworksMultilayer feedforward networks are universal approximators

21,248 Citations1989Kurt Hornik, Maxwell B. Stinchcombe +1 more

It is rigorously established that standard multilayer feedforward networks with as few as one hidden layer using arbitrary squashing functions are capable of approximating any Borel measurable function from one finite dimensional space to another to any desired degree of accuracy, provided sufficiently many hidden units are available.

Parallel ComputingGenetic algorithms and neural networks: optimizing connections and connectivity

749 Citations1990David S. Whitley, Timothy Starkweather +1 more

An overview of several different experiments applying genetic algorithms to neural network problems including optimizing the weighted connections in feed-forward neural networks using both binary and real-valued representations and using a genetic algorithm to discover novel architectures for neural networks that learn using error propagation are presented.

Combinations of genetic algorithms and neural networks: a survey of the state of the art

571 Citations2003J. David Schaffer, Darrell Whitley +1 more

An overview of this body of literature drawing out common themes and providing, where possible, the emerging wisdom about what seems to work and what does not is provided.

IEEE Transactions on Neural NetworksCooperative-competitive genetic evolution of radial basis function centers and widths for time series prediction

253 Citations1996Bruce A. Whitehead, T.D. Choate

In a radial basis function (RBF) network, the RBF centers and widths can be evolved by a cooperative-competitive genetic algorithm, and the feasibility of the resulting algorithm to evolve networks of Gaussian, inverse multiquadric, and thin-plate spline RBFs is demonstrated by predicting the Mackey-Glass time series.

Designing application-specific neural networks using the structured genetic algorithm

125 Citations2003Dipankar Dasgupta, D. R. McGregor

The empirical studies show that the SGA can efficiently determine the network size and topology along with the optimal set of connection weights appropriate for desired tasks, without using backpropagation or any other learning algorithm.

Using Genetic Algorithms to Improve Pattern Classification Performance

79 Citations1990Eric Chang, Richard P. Lippmann

The results suggest that genetic algorithms are becoming practical for pattern classification problems as faster serial and parallel computers are developed.

Hierarchical genetic fuzzy controller for a solar power plant

17 Citations2002Jiachen Ke, K.S. Tang +2 more

A unique hierarchical genetic algorithm (HGA) has been employed for the design and optimization of the fuzzy logic controller that not only fulfils the required performance, but also minimises the number of fuzzy membership functions and rules applied.

Proceedings of International Conference on Neural Networks (ICNN'96)Speech signal restoration using an optimal neural network structure

10 Citations2002Xiaoming Gao, S.J. Ovaska +1 more

An optimal neural network-based method for noisy speech restoration that uses a feedforward neural network with one hidden layer as a nonlinear predictive filter and applies the predictive minimum description length principle to determine the optimal number of input and hidden nodes.

Selecting an optimal neural network

5 Citations2002David B. Fogel

A derivative of Akaike's information criterion (AIC) is given and it is shown that this modification yields an information statistic which can be used to select a best network for binary classification problems objectively.