Auto-adapted Ant Colony Optimization Algorithm for Wavelet Network and Its Applications

In order to solve problems in wavelet network backward propagation, such as low-precision, slow learning process and easy convergence to the local minimum points, ant colony algorithm was modified. A wavelet network learning algorithm, which is based on modified auto-adapted ant colony algorithm, was put forward. Its application example of custom-made product cost estimation was given at last, which shows learning process and accuracy of the algorithm is better than others, and wavelet network training based on this algorithm has greater generality and better learning capacities