A Novel Manifold Regularized Online Semi-supervised Learning Algorithm

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 机器人应用与理论组

	A Novel Manifold Regularized Online Semi-supervised Learning Algorithm
	Ding, Shuguang; Xi, Xuanyang; Liu, Zhiyong; Qiao, Hong; Zhang, Bo
	2016
会议名称	23rd International Conference on Neural Information Processing (ICONIP)
会议日期	OCT 16-21, 2016
会议地点	Kyoto, JAPAN
关键词	Manifold regularization Online semi-supervised learning Lagrange dual problem
通讯作者	Liu, ZY
英文摘要	In this paper, we propose a novel manifold regularized online semi-supervised learning ((OSL)-L-2) model in an Reproducing Kernel Hilbert Space (RK-HS). The proposed algorithm, named Model-BasedOnline 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.
会议录	NEURAL INFORMATION PROCESSING, ICONIP 2016, PT I
内容类型	会议论文
源URL	[http://ir.ia.ac.cn/handle/173211/12836]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_机器人应用与理论组
推荐引用方式 GB/T 7714	Ding, Shuguang,Xi, Xuanyang,Liu, Zhiyong,et al. A Novel Manifold Regularized Online Semi-supervised Learning Algorithm[C]. 见:23rd International Conference on Neural Information Processing (ICONIP). Kyoto, JAPAN. OCT 16-21, 2016.