Back
We incorporate prior knowledge to construct nonlinear algorithms for invariant feature extraction and discrimination. Employing a unified framework in terms of a nonlinearized variant of the Rayleigh coefficient, we propose nonlinear generalizations of Fisher‘s discriminant and oriented PCA using support vector kernel functions. Extensive simulations show the utility of our approach.
@article{1844, title = {Constructing Descriptive and Discriminative Non-linear Features: Rayleigh Coefficients in Kernel Feature Spaces}, journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, abstract = {We incorporate prior knowledge to construct nonlinear algorithms for invariant feature extraction and discrimination. Employing a unified framework in terms of a nonlinearized variant of the Rayleigh coefficient, we propose nonlinear generalizations of Fisher‘s discriminant and oriented PCA using support vector kernel functions. Extensive simulations show the utility of our approach.}, volume = {25}, number = {5}, pages = {623-628}, organization = {Max-Planck-Gesellschaft}, school = {Biologische Kybernetik}, month = may, year = {2003}, author = {Mika, S. and R{\"a}tsch, G. and Weston, J. and Sch{\"o}lkopf, B. and Smola, AJ. and M{\"u}ller, K-R.}, doi = {10.1109/TPAMI.2003.1195996}, month_numeric = {5} }
More information