TY - GEN
T1 - A novel manifold regularized online semi-supervised learning algorithm
AU - Ding, Shuguang
AU - Xi, Xuanyang
AU - Liu, Zhiyong
AU - Qiao, Hong
AU - Zhang, Bo
N1 - Publisher Copyright:
© Springer International Publishing AG 2016.
PY - 2016
Y1 - 2016
N2 - In this paper, we propose a novel manifold regularized online semi-supervised learning (OS2L) model in an Reproducing Kernel Hilbert Space (RK-HS). The proposed algorithm, named Model-Based Online Manifold Regularization (MOMR), is derived by solving a constrained optimization problem, which is different from the stochastic gradient algorithm used for solving the online version of the primal problem of Laplacian support vector machine (LapSVM). Taking advantage of the convex property of the proposed model, an exact solution can be obtained iteratively by solving its Lagrange dual problem. Furthermore, a buffering strategy is introduced to improve the computational efficiency of the algorithm. Finally, the proposed algorithm is experimentally shown to have a comparable performance to the standard batch manifold regularization algorithm.
AB - In this paper, we propose a novel manifold regularized online semi-supervised learning (OS2L) model in an Reproducing Kernel Hilbert Space (RK-HS). The proposed algorithm, named Model-Based Online Manifold Regularization (MOMR), is derived by solving a constrained optimization problem, which is different from the stochastic gradient algorithm used for solving the online version of the primal problem of Laplacian support vector machine (LapSVM). Taking advantage of the convex property of the proposed model, an exact solution can be obtained iteratively by solving its Lagrange dual problem. Furthermore, a buffering strategy is introduced to improve the computational efficiency of the algorithm. Finally, the proposed algorithm is experimentally shown to have a comparable performance to the standard batch manifold regularization algorithm.
KW - Lagrange dual problem
KW - Manifold regularization
KW - Online semi-supervised learning
UR - https://www.scopus.com/pages/publications/84992592951
U2 - 10.1007/978-3-319-46687-3_66
DO - 10.1007/978-3-319-46687-3_66
M3 - 会议稿件
AN - SCOPUS:84992592951
SN - 9783319466866
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 597
EP - 605
BT - Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
A2 - Doya, Kenji
A2 - Ikeda, Kazushi
A2 - Lee, Minho
A2 - Hirose, Akira
A2 - Ozawa, Seiichi
A2 - Liu, Derong
PB - Springer Verlag
T2 - 23rd International Conference on Neural Information Processing, ICONIP 2016
Y2 - 16 October 2016 through 21 October 2016
ER -