Scaling the kernel function based on the separating boundary in input space: A data-dependent way for improving the performance of kernel methods.
Jiancheng SunXiaohe LiYong YangJianguo LuoYaohui BaiPublished in: Inf. Sci. (2012)
Keyphrases
- kernel function
- input space
- kernel methods
- data dependent
- reproducing kernel hilbert space
- support vector
- feature space
- support vector machine
- kernel matrix
- learning problems
- hyperplane
- kernel pca
- high dimensional feature space
- support vectors
- learning tasks
- kernel trick
- high dimensional
- generalization bounds
- positive definite
- linearly separable
- multiple kernel learning
- gaussian kernel
- kernel matrices
- energy functional
- kernel machines
- kernel learning
- loss function
- feature set
- active learning
- pattern recognition
- image segmentation
- computer vision
- machine learning