2008-07-09-bo08a.md

abstract	title	year	layout	series	publisher	issn	id	month	tex_title	firstpage	lastpage	page	order	cycles	bibtex_author	author	date	note	address	container-title	volume	genre	issued	pdf	extras
We propose a variable decomposition algorithm-greedy block coordinate descent (GBCD)-in order to make dense Gaussian process regression practical for large scale problems. GBCD breaks a large scale optimization into a series of small sub-problems. The challenge in variable decomposition algorithms is the identification of a sub-problem (the active set of variables) that yields the largest improvement. We analyze the limitations of existing methods and cast the active set selection into a zero-norm constrained optimization problem that we solve using greedy methods. By directly estimating the decrease in the objective function, we obtain not only efficient approximate solutions for GBCD, but we are also able to demonstrate that the method is globally convergent. Empirical comparisons against competing dense methods like Conjugate Gradient or SMO show that GBCD is an order of magnitude faster. Comparisons against sparse GP methods show that GBCD is both accurate and capable of handling datasets of 100,000 samples or more.	Greedy block coordinate descent for large scale Gaussian process regression	2008	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	bo08a	0	Greedy block coordinate descent for large scale Gaussian process regression	43	52	43-52	43	false	Bo, Liefeng and Sminchisescu, Cristian	given family Liefeng Bo given family Cristian Sminchisescu	2008-07-09	Reissued by PMLR on 09 October 2024.		Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence	R6	inproceedings	date-parts 2008 7 9	https://raw.githubusercontent.com/mlresearch/r6/main/assets/bo08a/bo08a.pdf