Avoiding weight-illgrowth: cascade correlation algorithm with local regularization
Qiang Wu, Kenji Nakayama · Proceedings of International Conference on Neural Networks (ICNN'97) · 2002
This paper investigates some possible problems of cascade correlation algorithm, one of which is the zigzag output mapping caused by weight-illgrowth of the adding hidden unit. Without doubt, it could lead to deterioration of the generalization, especially for regression problems. To solve this problem, we combine the cascade correlation algorithm with regularization theory. In addition, some new regularization terms are proposed in light of special cascade structure. Simulation has shown that regularization can smooth the zigzag output, so that the generalization is improved, especially for functional approximation.