Radial basis function process neural network training based on generalized frechet distance and GA-SA hybrid strategy

TL;DR

This paper introduces a novel training method for Radial Basis Function Process Neural Networks using a generalized Fréchet distance measure combined with a GA-SA hybrid optimization strategy, enhancing training efficiency and stability.

Contribution

It proposes a new optimization approach that integrates generalized Fréchet distance with a GA-SA hybrid algorithm for training RBF-PNNs, improving convergence and robustness.

Findings

Enhanced training efficiency demonstrated in experiments

Improved stability of the neural network training process

Effective global optimization of network parameters

Abstract

For learning problem of Radial Basis Function Process Neural Network (RBF-PNN), an optimization training method based on GA combined with SA is proposed in this paper. Through building generalized Fr\'echet distance to measure similarity between time-varying function samples, the learning problem of radial basis centre functions and connection weights is converted into the training on corresponding discrete sequence coefficients. Network training objective function is constructed according to the least square error criterion, and global optimization solving of network parameters is implemented in feasible solution space by use of global optimization feature of GA and probabilistic jumping property of SA . The experiment results illustrate that the training algorithm improves the network training efficiency and stability.