Sparse inversion for derivative of log determinant

Abstract

Algorithms for Gaussian process, marginal likelihood methods or restricted maximum likelihood methods often require derivatives of log determinant terms. These log determinants are usually parametric with variance parameters of the underlying statistical models. This paper demonstrates that, when the underlying matrix is sparse, how to take the advantage of sparse inversion---selected inversion which share the same sparsity as the original matrix---to accelerate evaluating the derivative of log determinant.

arXiv comment15

References in corpus (7)

• Scalable Log Determinants for Gaussian Process Kernel Learning
• Extension and optimization of the FIND algorithm: computing Green's and less-than Green's functions (with technical appendix)
• Bayesian Inference of Log Determinants
• Essential formulae for restricted maximum likelihood and its derivatives associated with the linear mixed models
• Learning with linear mixed model for group recommendation systems
• Simultaneous analysis of large INTEGRAL/SPI datasets: optimizing the computation of the solution and its variance using sparse matrix algorithms
• Computing log-likelihood and its derivatives for restricted maximum likelihood methods