Optimizing working sets for training support vector regressors by Newton's method.
Shigeo AbePublished in: IJCNN (2015)
Keyphrases
- high accuracy
- detection method
- similarity measure
- training process
- objective function
- support vector
- classification accuracy
- significant improvement
- clustering method
- support vector machine svm
- computational cost
- dynamic programming
- cost function
- pairwise
- computational complexity
- video sequences
- segmentation method
- multi class classification
- kernel methods
- data sets
- loss function
- support vector machine
- high dimensional
- face recognition