R (2000) Second-order learning algorithm with squared penalty term. Neural Comput 12:709–729

This paper compares three penalty terms with respect to the effi-ciency of supervised learning, by using first- and second-order learn-ing algorithms. Our experiments showed that for a reasonably ade-quate penalty factor, the combination of the squared penalty term and the second-order learning algorithm drastically improves the convergence performance more than 20 times over the other com-binations, at the same time bringing about a better generalization performance.