Show simple item record

Nonlinear System Identification with Composite Relevance Vector Machines

dc.contributor.authorCamps Valls, Gustavo
dc.contributor.authorMartínez Ramón, Manel
dc.contributor.authorRojo-Álvarez, José Luis
dc.contributor.authorMuñoz Marí, Jordi
dc.description.abstractNonlinear system identification based on relevance vector machines (RVMs) has been traditionally addressed by stacking the input and/or output regressors and then performing standard RVM regression. This letter introduces a full family of composite kernels in order to integrate the input and output information in the mapping function efficiently and hence generalize the standard approach. An improved trade-off between accuracy and sparsity is obtained in several benchmark problems. Also, the RVM yields confidence intervals for the predictions, and it is less sensitive to free parameter
dc.relation.ispartofseriesIEEE Signal Processing Letterses
dc.rightsAtribución-NoComercial-SinDerivadas 3.0 España*
dc.titleNonlinear System Identification with Composite Relevance Vector Machineses
dc.subject.unesco3325 Tecnología de las Telecomunicacioneses
dc.description.departamentoTeoría de la Señal y Comunicaciones

Files in this item

This item appears in the following Collection(s)

Show simple item record

Atribución-NoComercial-SinDerivadas 3.0 EspañaExcept where otherwise noted, this item's license is described as Atribución-NoComercial-SinDerivadas 3.0 España