Conference paper
Column-generation boosting methods for mixture of kernels
Jinbo Bi, Tong Zhang, et al.
KDD 2004
In this article, we study leave-one-out style cross-validation bounds for kernel methods. The essential element in our analysis is a bound on the parameter estimation stability for regularized kernel formulations. Using this result, we derive bounds on expected leave-one-out cross-validation errors, which lead to expected generalization bounds for various kernel algorithms. In addition, we also obtain variance bounds for leave-one-out errors. We apply our analysis to some classification and regression problems and compare them with previous results.
Jinbo Bi, Tong Zhang, et al.
KDD 2004
Rie Johnson, Tong Zhang
JMLR
Tong Zhang
IEEE Trans. Inf. Theory
Tong Zhang
Neural Computation