Data-independent Random Projections from the feature-space of the homogeneous polynomial kernel.

[EN]Performing a Random Projection from the feature space associated to a kernel function may be impor- tant for two main reasons. As a consequence of the Johnson–Lindestrauss lemma, the resulting low- dimensional representation will preserve most of the structure of data in the kernel feature space...

Descripción completa

Detalles Bibliográficos
Autores: López Sánchez, Daniel, González Arrieta, María Angélica, Corchado Rodríguez, Juan Manuel
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2018
País:España
Institución:Universidad de Salamanca (USAL)
Repositorio:GREDOS. Repositorio Institucional de la Universidad de Salamanca
OAI Identifier:oai:gredos.usal.es:10366/157256
Acceso en línea:http://hdl.handle.net/10366/157256
Access Level:acceso abierto
Palabra clave:Random Projection
Homogeneous polynomial kernel
Nonlinear dimensionality reduction
1203.17 Informática
Descripción
Sumario:[EN]Performing a Random Projection from the feature space associated to a kernel function may be impor- tant for two main reasons. As a consequence of the Johnson–Lindestrauss lemma, the resulting low- dimensional representation will preserve most of the structure of data in the kernel feature space and (2) an efficient linear classifier trained on transformed data might approximate the accuracy of its nonlinear counterparts. In this paper, we present a novel method to perform Random Projections from the feature space of homogeneous polynomial kernels. As opposed to other kernelized Random Projection propos- als, our method focuses on a specific kernel family to preserve some of the beneficial properties of the original Random Projection algorithm (e.g. data independence and efficiency). Our extensive experimental results evidence that the proposed method efficiently approximates a Random Projection from the kernel feature space, preserving pairwise distances and enabling a boost on linear classification accuracies.