Generalized Back-Propagation Algorithm with Weight Evolution for Neural Networks

摘要

The conventional back-propagation algorithm will always get stuck into local minima and converge very slowly. Other fast algorithms can increase the convergence speed however they still sometimes converge to local solution. We introduce a new algorithm with the use of weight evolution into the generalized back-propagation method. The new algorithm further improves the convergence rate and the global convergence capability. It ensures the convergence to a global minimum in a compact region of a weight vector space.