Geodesic
Application of block low-rank matrices in Gaussian processes for regression
Numerical methods and programming, Tome 18 (2017) no. 3, pp. 214-220.

Voir la notice de l'article provenant de la source Math-Net.Ru

The Gaussian processes for regression are considered. During simulation of correlated noises using the Gaussian processes, the main difficulty is the computation of the posterior mean and dispersion of the prediction. This computation requires the inversion of the dense covariance matrix of order $n$, where $n$ is the sample size. In addition, for the likelihood evaluation we need to compute the logarithm of the determinant of the dense covariance matrix, which is also a time-consuming problem. A new method for the fast computation of the covariance matrix logarithm is proposed. This method is based on the approximation of this matrix by a sparse matrix. The proposed method appears to be time efficient compared to the HODLR (Hierarchically Off-Diagonal Low-Rank) method and the traditional dense method.
Keywords: Gaussian processes, $\mathcal{H}^2$ matrix, Cholesky factorization.
Mots-clés : sparse matrix 
@article{VMP_2017_18_3_a2,
     author = {D. A. Sushnikova},
     title = {Application of block low-rank matrices in {Gaussian} processes for regression},
     journal = {Numerical methods and programming},
     pages = {214--220},
     publisher = {mathdoc},
     volume = {18},
     number = {3},
     year = {2017},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2017_18_3_a2/}
}
TY  - JOUR
AU  - D. A. Sushnikova
TI  - Application of block low-rank matrices in Gaussian processes for regression
JO  - Numerical methods and programming
PY  - 2017
SP  - 214
EP  - 220
VL  - 18
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2017_18_3_a2/
LA  - ru
ID  - VMP_2017_18_3_a2
ER  - 
%0 Journal Article
%A D. A. Sushnikova
%T Application of block low-rank matrices in Gaussian processes for regression
%J Numerical methods and programming
%D 2017
%P 214-220
%V 18
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2017_18_3_a2/
%G ru
%F VMP_2017_18_3_a2
D. A. Sushnikova. Application of block low-rank matrices in Gaussian processes for regression. Numerical methods and programming, Tome 18 (2017) no. 3, pp. 214-220. http://geodesic.mathdoc.fr/item/VMP_2017_18_3_a2/