Relation to Existing Work Comparison with RGP [Chu et al.,2007]and XGP [Silva et al.,2008] 。RGP and XGP: Learn only one GP. p(BZ)is itself a prediction function with B being a vector of function values for all input points. The learned kernel,which is the covariance matrix of the posterior distribution p(BZ).is (K-1+)-1 in RGP and (K+n)in XGP. where n is a kernel matrix capturing the link information. LWP: Learn multiple(g)GPs. Treat A=BB'as the learned kernel matrix. 4口4日+1立4至卡三只0 Li,Zhang and Yeung (CSE.HKUST) LWP A1 STATS200915/23Relation to Existing Work Comparison with RGP [Chu et al. , 2007] and XGP [Silva et al. , 2008] RGP and XGP: Learn only one GP. p(B|Z) is itself a prediction function with B being a vector of function values for all input points. The learned kernel, which is the covariance matrix of the posterior distribution p(B|Z), is (K−1 + Π −1 ) −1 in RGP and (K + Π) in XGP, where Π is a kernel matrix capturing the link information. LWP: Learn multiple (q) GPs. Treat A = BB0 as the learned kernel matrix. Li, Zhang and Yeung (CSE, HKUST) LWP AISTATS 2009 15 / 23