SciELO - Scientific Electronic Library Online

 
vol.29 número4Predicción de Riesgos Psicosociales en Docentes de Colegios Públicos Colombianos utilizando Técnicas de Inteligencia ArtificialInfluencia del Y Plus en el Valor del Esfuerzo Cortante de Pared a través Simulaciones empleando Dinámica Computacional de Fluidos índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google

Compartir


Información tecnológica

versión On-line ISSN 0718-0764

Resumen

HOYOS, Juan P.  y  JOJOA, Pablo E.. A Review of Low Rank and Sparse Matrix Estimators. Inf. tecnol. [online]. 2018, vol.29, n.4, pp.281-290. ISSN 0718-0764.  http://dx.doi.org/10.4067/S0718-07642018000400281.

A selective review of the most recent developments in the estimation of the covariance matrix with a number of samples smaller than the ambient dimension is presented. In particular, estimators are considered for low rank and sparse structures, paying special attention to the algorithms most used in practice. Estimators that resort to the use of classical techniques such as thresholds and singular-value decomposition (SVD), as well as newest estimators based on random matrices and convex optimization are presented. One of the main conclusions of the study is that, in spite of the great advances in the development of new estimators, issues related to the computation time of the convex programs that use the estimators have been little explored. Additionally, lack of works on the development of estimators for matrices with jointly sparse and low rank structure is observed

Palabras clave : covariance matrix; low rank; sparse; sketch; estimators; random matrices.

        · resumen en Español     · texto en Español     · Español ( pdf )