Sacrificial Learning in Nonlinear Perceptrons

Simple item page
dc.creatorLuo, Peixun
dc.creatorWong, K. Y. Michael
dc.date2000-06-13
dc.date.accessioned2026-07-07T02:37:54Z
dc.date.available2026-07-07T02:37:54Z
dc.descriptionUsing the cavity method we consider the learning of noisy teacher-generated examples by a nonlinear student perceptron. For insufficient examples and weak weight decay, the activation distribution of the training examples exhibits a gap for the more difficult examples. This illustrates that the outliers are sacrificed for the overall performance. Simulation shows that the picture of the smooth energy landscape cannot describe the gapped distributions well, implying that a rough energy landscape may complicate the learning process.
dc.description7 pages, 4 figures
dc.identifierhttps://arxiv.org/abs/cond-mat/0006206
dc.identifierhttp://arxiv.org/abs/cond-mat/0006206
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/16472
dc.subjectDisordered Systems and Neural Networks
dc.titleSacrificial Learning in Nonlinear Perceptrons
dc.typetext

Files

Collections

arXiv