Timeline for Problem with singular covariance matrices when doing Gaussian process regression

6 events

when toggle format	what		by	license	comment
Jan 16, 2012 at 11:04	comment	added	Dikran Marsupial		The key trick to limiting numerical instability is on page 45 (equation 3.26).
Jan 16, 2012 at 9:56	comment	added	Andreas		"there is a good discussion of this in Rasmussen and Williams book" - May i ask where exactly
Jan 13, 2012 at 15:24	comment	added	Dikran Marsupial		The matrix is often ill-conditioned, so there are indeed a few tricks for inverting it more reliably, there is a good discussion of this in Rasmussen and Williams book. However, I have found that I normally only run into problems when model selection tries to make a very bland RBF covariance to model an essentially linear decision boundary, so you could argue it was model mis-specification? I haven't used GP regression very much, so it is hard to know whether it crops up more often there.
Jan 13, 2012 at 10:35	comment	added	Andreas		I read in some publications about estimating the inverse for singular matrices. Is that a standard problem in gaussian process regression or why is there so much literature about numerical problems in the covariance matrices.
Jan 13, 2012 at 10:31	vote	accept	Andreas
Jan 13, 2012 at 10:02	history	answered	Dikran Marsupial	CC BY-SA 3.0