Research Project
Grant-in-Aid for Young Scientists (B)
The 1-norm soft margin optimization is a popular formulation for obtaining sparse classifiers. We propose a new boosting algorithm based on linear programming. Our algorithm can take advantage of the sparsity of the solution of the underlying optimization problem. In preliminary experiments, our algorithm outperforms a state-of-the-art LP solver and LPBoost.
All 2011 2010 2009 Other
All Journal Article (10 results) (of which Peer Reviewed: 5 results) Presentation (20 results) Remarks (3 results)
Proc.of the 2nd Asian Conference on Machine Learning (ACML 2010)
Pages: 31-46
Proc.of the 13th International Conference on Discovery Science
Pages: 132-144
Proceedings of the 2nd Asian Conference Conference on Machine Learning
Proceedings of the 13th International Conference on Discovery Science
Pages: 132-143
Proc.of 12th International Conference on Discovery Science
Pages: 401-408
120006654515
Neural Computation vol.21, No.5
Pages: 1459-1484
人工知能学会論文誌 vol.24, No.1
Pages: 191-202
130000098282
Neural Computation 21(5)
人工知能学会論文誌 24(1)
Proceedings of the 12th International Conference on Discovery Science
http://www.i.kyushu-u.ac.jp/~hatano/