Hierarchical Learning Algorithm for the Beta Basis Function Neural Network
This work addresses the problem of designing neural networks for function approximation, but it appears incremental as it combines existing methods (Genetic Algorithm and gradient optimization) without clear broad impact.
The paper tackles the design of Beta Basis Function Neural Networks (BBFNN) by introducing a two-level hierarchical learning algorithm (HLABBFNN) that uses a Genetic Algorithm at the upper level and gradient optimization at the lower level to optimize parameters like width, centers, and Beta form, and demonstrates its effectiveness for approximating non-linear functions, though no concrete numerical results are provided.
The paper presents a two-level learning method for the design of the Beta Basis Function Neural Network BBFNN. A Genetic Algorithm is employed at the upper level to construct BBFNN, while the key learning parameters :the width, the centers and the Beta form are optimised using the gradient algorithm at the lower level. In order to demonstrate the effectiveness of this hierarchical learning algorithm HLABBFNN, we need to validate our algorithm for the approximation of non-linear function.