Fletcher R., Johnson T., "On the stability of nullspace methods for KKT systems," SIAM J. Matrix Anal. Appl., Vol.18, No.4, 938-958, 1997. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Multiclass Least Squares Support Vector Machines - Suykens, Vandewalle (1999) (4 citations) (Correct)
.... : L w = 0 w = P N k=1 ff k y k (x k ) L b = 0 P N k=1 ff k y k = 0 L ek = 0 ff k = fle k L ff k = 0 y k [w T (x k ) b] Gamma 1 e k = 0 (7) for k = 1; N can be written after elimination of w and e as the linear system [3, 4]: 0 Y T Y ZZ T fl Gamma1 I b ff = 0 1 (8) where Z = x 1 ) T y 1 ; xN ) T yN ] Y = y 1 ; yN ] 1 = 1; 1] e = e 1 ; e N ] ff = ff 1 ; ff N ] Mercer s condition is applied to the matrix Omega = ZZ T with Omega kl = y k y l (x k ) ....
Fletcher R., Johnson T., "On the stability of nullspace methods for KKT systems," SIAM J. Matrix Anal. Appl., Vol.18, No.4, 938-958, 1997.
Least Squares Support Vector Machine Classifiers.. - Suykens, Lukas.. (1999) (15 citations) (Correct)
....been proposed, which is related to the LS version for function estimation reported in [10] In this LS SVM version one finds the solution by solving a linear system instead of quadratic programming. This is due to the use of equality instead of inequality constraints in the problem formulation. In [2, 5, 13] such linear systems have been called augmented systems or Karush Kuhn Tucker (KKT) systems and their numerical stability has been investigated. In this paper we present an iterative solution to LS SVM s based on the conjugate gradient method [7] This method enables solving large scale ....
Fletcher R., Johnson T., "On the stability of nullspace methods for KKT systems," SIAM J. Matrix Anal. Appl., Vol.18, No.4, 938-958, 1997.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC