Projective Unsupervised Flexible Embedding with Optimal Graph

Wei Wang, Yan Yan, Feiping Nie, Xavier Pineda, Shuicheng Yan and Nicu Sebe

Abstract

Graph based dimensionality reduction techniques have been successfully applied to clustering and classification tasks. The fundamental basis of these algorithms is the constructed graph which dominates their performance. Usually, the graph is defined by the input affinity matrix. However, the affinity matrix is sub-optimal for dimension reduction as there is much noise in the data. To address this issue, we propose the projective unsupervised flexible embedding with optimal graph (PUFE-OG) model. We build an optimal graph by adjusting the affinity matrix. To tackle the out-of-sample problem, we employ a linear regression term to learn a projection matrix. The optimal graph and projection matrix are jointly learned by integrating the manifold regularizer and regression residual into a unified model. An efficient algorithm is derived to solve the challenging model. The experimental results on several public benchmark datasets demonstrate that the presented PUFE-OG outperforms other state-of-the-art methods.

Session

Posters 2

Files

Extended Abstract (PDF, 41K)
Paper (PDF, 333K)
Supplemental Materials (ZIP, 118K)

DOI

10.5244/C.30.100
https://dx.doi.org/10.5244/C.30.100

Citation

Bibtex

        @inproceedings{BMVC2016_100,
        	title={Projective Unsupervised Flexible Embedding with Optimal Graph},
        	author={Wei Wang, Yan Yan, Feiping Nie, Xavier Pineda, Shuicheng Yan and Nicu Sebe},
        	year={2016},
        	month={September},
        	pages={100.1-100.12},
        	articleno={100},
        	numpages={12},
        	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
        	publisher={BMVA Press},
        	editor={Richard C. Wilson, Edwin R. Hancock and William A. P. Smith},
        	doi={10.5244/C.30.100},
        	isbn={1-901725-59-6},
        	url={https://dx.doi.org/10.5244/C.30.100}
        }