Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4947178 | Neurocomputing | 2017 | 7 Pages |
Abstract
In this paper, a smooth function is constructed to approximate the nonsmooth output of maxâ-minâ fuzzy neural networks (FNNs) and its approximation is also presented. In place of the output of maxâ-minâ FNNs by its smoothing approximation function, the error function, defining the discrepancy between the actual outputs and desired outputs of maxâ-minâ FNNs, becomes a continuously differentiable function. Then, a smoothing gradient decent-based algorithm with Armijo-Goldstein step size rule is formulated to train maxâ-minâ FNNs. Based on the existing convergent result, the convergence of our proposed algorithm can easily be obtained. Furthermore, the proposed algorithm also provides a feasible procedure to solve fuzzy relational equations with maxâ-minâcomposition. Finally, some numerical examples are implemented to support our results and demonstrate that the proposed smoothing algorithm has better learning performance than other two gradient decent-based algorithms.
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
Long Li, Zhijun Qiao, Yan Liu, Yuan Chen,