Optimization of feed-forward neural networks based on artificial fish-swarm algorithm
14
Citation
0
Reference
20
Related Paper
Citation Trend
Abstract:
Artificial Fish-swarm Algorithm(AFSA) is a novel optimizing method proposed lately.An Artificial Fish-swarm Algorithm(AFSA) for the optimization of feed-forward neural networks and a model based on this method were presented for the first time here.Compared with the Back-propagation Algorithm added momentum,the Evolve Algorithm and the Simulated Anncaling Algorithm,optimization result of feed-forward neural networks by AFSA demonstrates that AFSA has a strong robustness and good global astringency.AFSA is also proved to be insensitive to initial values.Keywords:
Robustness
Optimization algorithm
Backpropagation
Cite
In view of the shortcomings of AFSA(artificial fish swarm algorithm),the self-adjustment strategy on visual range and step size is introduced,so as to increase search efficiency and convergence rate.The improved AFSA may synchronously determine the parameters initiation value and hidden layer nodes number in search space.The simulation is given to illustrate the effectiveness of the method.
Optimization algorithm
Cite
Citations (0)
Sinter tumbler strength is an important parameter in the sintering process, and has an important influence on the performance of finished sinter. Artificial fish swarm algorithm have good ability to acquire the global performance, the neural network has strong nonlinear ability and local optimization performance,; AFSA+BP algorithm combined with artificial fish swarm algorithm and BP algorithm, realizes the complementary artificial fish swarm algorithm global search capability and BP algorithm's local optimization combination of performance, an artificial fish swarm neural results show that the network combination algorithm, it is shown that comparing with the traditional BP neural network forecasting method，the presented forecasting method has better adaptive ability and can give better forecasting results.The artificial fish—swarm algorithm network is trained and checked with the actual production data．this algorithm has strong generalization capability, predictive accuracy improved significantly, and speed up the convergence rate, provides an effective method for strength prediction. Which be used for off-line learning and prediction, a good basis for the online application.
Cite
Citations (0)
After analyzing the disadvantages of AFSA,this paper introduced best-step operator and refined the prey behavior.It developed an improved artificial fish-swarm algorithm for the RBF neural network and a model based on this method.Finally applied the new algorithm to the problem of expression recognition.The research indicates that the new algorithm has some advantages in terms of convergence performance,recognition rate and so on.
Cite
Citations (0)
To describe the complex nonlinear characteristics of a system accurately, a Wavelet Neural Network (WNN) identification model based on Artificial Fish Swarm (AFS) algorithm is proposed. In the identification model, AFS algorithm is introduced to optimize the parameters combination of the network for the satisfactory WNN model. The simulation shows that, the proposed method is a good nonlinear identification capability, and is feasible to identify the nonlinear system.
Identification
Nonlinear system identification
Cite
Citations (0)
Cite
Citations (0)
Artificial fish-swarm algorithm(AFSA) is a nove1 optimizing method proposed lately.An Artificial Fish-swarm Algorithm(AFSA) for the RBF neural networks and a model based on this method were presented of the first time here.Compared with the Back-propagation Algorithm added momentum and the RBF Algorithm,optimization result of RBF neural networks by AFSA demonstrates that AFSA has a strong robustness and good global astringency.AFSA is also proved to be initial values.
Robustness
Cite
Citations (2)
Artificial fish swarm algorithm (AFSA) is a global optimization method proposed recently. After analyzing the disadvantages of AFSA, this paper introduced best-step operator and refined the prey behavior. An improved artificial fish-swarm algorithm for the RBF neural network and a model based on this method is developed. Finally the new algorithm is applied to the problem of expression recognition. The research indicates that the new algorithm has some advantages in terms of convergence performance, recognition rate and so on.
Cite
Citations (9)
In order to solve problems such as initial weights are difficult to be determined, training results are easy to trap in local minima in optimization process of PID neural network parameters by traditional BP algorithm, this paper proposed a new method based on improved artificial fish algorithm for parameters optimization of PID neural network. This improved artificial fish algorithm uses a composite adaptive artificial fish algorithm based on optimal artificial fish and nearest artificial fish to train network weights parameters of PID neural network. By comparing food consistence in preying behavior to adaptively select vision and step of artificial fish, this method overcomes shortcomings such as slow convergence speed, low optimization accuracy of basic artificial fish algorithm. Simulations of PID neural network system whose parameters are trained respectively through BP algorithm and improved artificial fish algorithm are conducted respectively in the MATLAB environment. The simulation result shows that the PID neural network control system whose parameters are trained by the improved artificial fish algorithm has a better control effect, especially for nonlinear systems
Cite
Citations (2)
In order to improve the modeling efficiency of RBF neural network, an Artificial Fish Swarm Algorithm (AFSA) training algorithm with an adaptive mechanism is proposed. In the training algorithm, the search step size and visible domain of AFSA algorithm can be adjusted dynamically according to the convergence characteristics of artificial fish swarm, and then the improved AFSA algorithm is used to optimize the parameters of RBF neural network. The example shows that, the proposed model is a better approximation performance for the nonlinear function.
Cite
Citations (3)
Artificial fish-swarm algorithm is a realization model of the swarm intelligence optimization algorithm.It uses the optimization model of imitated nature fish for feeding from top to bottom, clusters and rear, local optimization by individual fish, achieve the purpose of global optimal values highlighted in the groups.RBFNN based on the AFSA can accurately find the optimal solution quickly and ensure the diversity of artificial fish.It is easier to find the global optimal point of optimal fish.This design uses second-order pendulum as a controlled object, using artificial fish swarm algorithm applied to the neural network training algorithms, building design of RBF Neural networks control module , verifing by Matlab simulation of actual control controller performance.
Cite
Citations (2)