An efficient SMO-like algorithm for multiclass SVM
Tóm tắt
Starting from a reformulation of Cramer and Singer (see Journal of Machine Learning Research, vol.2, p.265-92, Dec. 2001) multiclass kernel machine, we propose a sequential minimal optimization (SMO) like algorithm for incremental and fast optimization of the Lagrangian. The proposed formulation allowed us to define very effective new pattern selection strategies which lead to better empirical results.
Từ khóa
#Support vector machines #Kernel #Support vector machine classification #Lagrangian functions #PrototypesTài liệu tham khảo
platt, 1998, Fast training of Support Vector Machines using sequential minimal optimization, Advances in Kernel Methods - Support Vector Learning
dietterich, 1995, Solving multiclass learning problems via error correcting output codes, The Journal of Artificial Intelligence Research, 2, 263, 10.1613/jair.105
vapnik, 1998, Statistical Learning Theory
10.1023/A:1007614523901
crammer, 2001, On the Algorithmic Implementation of Multiclass Kernel-based Machines, Journal of Machine Learning Research, 2, 265
aiolli, 2001, A re-weighting strategy for improving margins, Artificial Intelligence Journal, 137 1?2, 197