ε-tube based Pattern Selection for Support Vector Machines

@inproceedings{Kim2005tubeBP,
  title={ε-tube based Pattern Selection for Support Vector Machines},
  author={Dongil Kim and Sungzoon Cho},
  year={2005}
}
The training time complexity of Support Vector Regression (SVR) is O(N). Hence, it takes long time to train a large dataset. In this paper, we propose a pattern selection method to reduce the training time of SVR. With multiple bootstrap samples, we estimate ε-tube. Probabilities are computed for each pattern to fall inside ε-tube. Those patterns with higher probabilities are selected stochastically. To evaluate the new method, the experiments for 4 datasets have been done. The proposed method… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

Citations

Publications citing this paper.
Showing 1-4 of 4 extracted citations

On Selecting Effective Patterns for Fast Support Vector Regression Training

IEEE Transactions on Neural Networks and Learning Systems • 2018
View 5 Excerpts
Highly Influenced

Tensor rank selection for multimedia analysis

J. Visual Communication and Image Representation • 2015
View 1 Excerpt

Pattern Selection for Support Vector Regression based on Sparseness and Variability

The 2006 IEEE International Joint Conference on Neural Network Proceedings • 2006

References

Publications referenced by this paper.
Showing 1-10 of 10 references

A heuristic training for support vector regression

Neurocomputing • 2004
View 9 Excerpts
Highly Influenced

Pattern Selection for Support Vector Regression based on Sparseness and Variability

The 2006 IEEE International Joint Conference on Neural Network Proceedings • 2006
View 1 Excerpt

The Nature of Statistical Learning Theory

Statistics for Engineering and Information Science • 2000
View 1 Excerpt

Mozer, M

H. Drucker, C.J.C. Burges, +4 authors In
C., Jordan, M. I., Petsche, T.(eds.): Advances in Neural Information Processing System 9, MIT Press, Cambridge, MA • 1997
View 2 Excerpts

Similar Papers

Loading similar papers…