Latent Wishart Processes Model Fommulaton Relational Data 0{(x1,y,zk)1i,k=1,,n} (月 21 (x yi) 2a=0 (xyi) 2微=1 x y) xi=(xi1,...,Xip)':input feature vector for instance i yi:label for instance i ozik =1 if there exists a link between xi and xk;0 otherwise. zik zki and zi =0.Z=[zik]i.k=1. 4口4日+1立4至卡三)风0 Li,Zhang and Yeung (CSE,HKUST) LWP A1 STATS20099/23Latent Wishart Processes Model Formulation Relational Data {(xi , yi , zik ) | i, k = 1, . . . , n} (xi , yi )  (xj , yj )  (xk, yk)  (xl , yl )  zik = 1  zij = 1  zil = 0  xi = (xi1, . . . , xip) 0 : input feature vector for instance i yi : label for instance i zik = 1 if there exists a link between xi and xk ; 0 otherwise. zik = zki and zii = 0. Z = [zik ] n i,k=1. Li, Zhang and Yeung (CSE, HKUST) LWP AISTATS 2009 9 / 23