613 search results - page 32 / 123 » The Pre-Image Problem in Kernel Methods |
JMLR
2006
13 years 7 months ago
2006
While classical kernel-based learning algorithms are based on a single kernel, in practice it is often desirable to use multiple kernels. Lanckriet et al. (2004) considered conic ...
AAAI
2010
13 years 9 months ago
2010
Multiple Kernel Learning (MKL) can be formulated as a convex-concave minmax optimization problem, whose saddle point corresponds to the optimal solution to MKL. Most MKL methods e...
ESANN
2004
13 years 9 months ago
2004
The problem of model selection for support vector machines (SVMs) is considered. We propose an evolutionary approach to determine multiple SVM hyperparameters: The covariance matr...
ICML
2002
IEEE
14 years 8 months ago
2002
IEEE
Semi-structured data such as XML and HTML is attracting considerable attention. It is important to develop various kinds of data mining techniques that can handle semistructured d...
NIPS
2003
13 years 9 months ago
2003
We propose a novel method of dimensionality reduction for supervised learning. Given a regression or classification problem in which we wish to predict a variable Y from an expla...