Feasible direction decomposition algorithms for training support vector machines

Machine Learning

Volumn 46, Issue 1-3, 2002, Pages 315-349

  Laskov, Pavel ^a  

^a University of Delaware   (United States)

Author keywords

Decomposition algorithms; Methods of feasible directions; Support vector machines; Training; Working set selection

Indexed keywords

COMPUTATIONAL COMPLEXITY; CONVERGENCE OF NUMERICAL METHODS; HEURISTIC METHODS; LAGRANGE MULTIPLIERS; LEARNING ALGORITHMS; MATRIX ALGEBRA; PATTERN RECOGNITION; QUADRATIC PROGRAMMING; REGRESSION ANALYSIS;

DIRECTION ALGORITHMS; FEASIBLE DIRECTION METHOD; SUPPORT VECTOR MACHINES; TRAINING ALGORITHMS; WORKING SET SELECTION;

LEARNING SYSTEMS;

EID: 0036158636     PISSN: 08856125     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1012479116909     Document Type: Article

References (24)

1. The analysis of decomposition methods for support vector machines
2. Support vector machines for large-scale regression problem
3. A simple decomposition method for support vector machines
4. Making large-scale support vector machine learning practical
5. Solving the quadratic problem arising in support vector classification
6. Improvements to Platt's SMO algorithm for SVM classifier design
7. An improved decomposition algorithm for regression support vector machines
8. Massive support vector regression
9. Successive overrelaxation for support vector machines
10. Support vector machines: Training and applications
11. Improved training algorithm for support vector machines
12. Fast training of support vector machines using sequential minimal optimization
13. Support vector learning
14. Advances in kernel methods - Support vector learning
15. Improvements to SMO algorithm for SVM regression
16. Learning with kernels
17. A tutorial on support vector regression