Inductive manifold learning using structured support vector machine

作者：

Highlights：

• We propose a general framework for out-of-sample extensions for any manifold learning methods.

• We learn a mapping from original space to its manifolds by using structured SVM.

• Experiments on several datasets show that the proposed method outperforms the existing ones.

摘要

Highlights•We propose a general framework for out-of-sample extensions for any manifold learning methods.•We learn a mapping from original space to its manifolds by using structured SVM.•Experiments on several datasets show that the proposed method outperforms the existing ones.

论文关键词：Dimensionality reduction,Manifold learning,Out-of-sample extension,Structured SVM

论文评审过程：Received 23 November 2012, Revised 28 June 2013, Accepted 16 July 2013, Available online 29 July 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.07.011