Neural-network learning algorithm based on particle swarm optimization

Abstract

The invention discloses a neural-network learning algorithm based on particle swarm optimization. Firstly, data are acquired to be used as training sample data of an RBF (radial-basis function) neural-network; subtractive clustering is carried out on the training sample data, and the number of basis-function centers is determined; a particle swarm is initialized; fitness of each particle in the particle swarm is processed, the current fitness of the particle is compared with historical optimal fitness, and pid is updated if the current fitness of the particle is better; the fitness of each particle is compared with fitness of a best position experienced by the population, and pgd is updated if the fitness of the particle is better; velocity and positions of particles are repeatedly adjusted until a requirement is met; values of a best position experienced by the entire population are used as parameters of the RBF neural-network after decoding, and training, accuracy verification and prediction of the neural network are carried out; and operations are stopped. The method has the advantage of enabling performance of the RBF neural-network to be better through introducing a particle swarm optimization algorithm of neural-network improvement.

Description

Technical field [0001] The invention belongs to the technical field of neural network algorithms, and relates to a neural network learning algorithm based on particle swarm optimization. Background technique [0002] The particle swarm optimization algorithm is an evolutionary calculation method based on swarm intelligence proposed by Dr. Eberhart and Dr. Kennedy in 1995. Compared with other evolutionary algorithms, it has simple, easy to implement and powerful global optimization capabilities. Therefore, with the development of society Progress, particle swarm algorithm has been greatly developed, and has been widely used in the fields of function optimization and neural network training. Compared with genetic algorithm, particle swarm optimization algorithm not only has global optimization ability, but also has excellent local optimization ability. It is a new optimization algorithm based on swarm intelligence and is more suitable for computer programming. [0003] Radial-basis ...

AI Technical Summary

Problems solved by technology

The orthogonal optimization algorithm can determine the optimal number of hidden layer nodes and the output weight of the network by itself, but the number of initial center points is too large, and data pathological phenomena will occur. When the selected orthogonal vectors exceed a certain number, further selection of orthogonal vectors The process of the vector will be difficult; the recursive Givens transformation algorithm solves the data ill-conditioned problem in the recursive least squares method, but the amount of calculation is too large; the method of determining the center of the basis function also has a genetic algorithm, which has a better effect, but it is relatively difficult to achieve Complex; K-means is often used to determine the center point of the odd function under the premise that the number of base function centers is determined, but it may be over-learning due to the excessive number of hidden nodes obtained by the clustering algorithm

Images