LS-based training algorithm for neural networks

A new training algorithm is presented as a faster alternative to the backpropagation method. The new approach is based on the solution of a linear system at each step of the learning phase. The squared error at the output of each layer before the nonlinearity is minimised on the entire set of the learning patterns by a block Least squares algorithm. The optimal weights for each layer are then computed by using the Singular Value Decomposition technique. The simulation results have shown considerable improvements from the point of view of both the accuracy and the speed of convergence.