Sketches by MoSSaRT: Representative selection from manifolds with gross sparse corruptions
作者：
Highlights：
• A reproduction profile encodes the data relations of grossly corrupted manifold structures.
• Approximate feature maps emulate a desired feature mapping associated with a RKHS.
• Scalable and parallelizable ADMM-based algorithm has nearly linear complexity in the data size.
• The proxy objective function induced by the approximate features converges exponentially fast.
• The representatives are vertices of the symmetrized convex hull of the data in a transformed space.
摘要
•A reproduction profile encodes the data relations of grossly corrupted manifold structures.•Approximate feature maps emulate a desired feature mapping associated with a RKHS.•Scalable and parallelizable ADMM-based algorithm has nearly linear complexity in the data size.•The proxy objective function induced by the approximate features converges exponentially fast.•The representatives are vertices of the symmetrized convex hull of the data in a transformed space.
论文关键词：Representative selection,Gross sparse corruption,Manifold learning,Reproducing kernel Hilbert spaces
论文评审过程：Received 27 August 2020, Revised 29 May 2021, Accepted 25 November 2021, Available online 27 November 2021, Version of Record 5 December 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.108454
