Discriminative clustering using regularized subspace learning

作者:

Highlights:

• A regularized subspace learning method for discriminative clustering is proposed.

• A similarity-based subspace learning formulation is used for the optimization.

• Robust representations optimized for clustering tasks are learned.

• The proposed method is extensively evaluated using four datasets.

摘要

•A regularized subspace learning method for discriminative clustering is proposed.•A similarity-based subspace learning formulation is used for the optimization.•Robust representations optimized for clustering tasks are learned.•The proposed method is extensively evaluated using four datasets.

论文关键词:Discriminative clustering,Subspace learning,Unsupervised learning

论文评审过程:Received 7 April 2018, Revised 3 May 2019, Accepted 21 July 2019, Available online 22 July 2019, Version of Record 26 July 2019.

论文官网地址:https://doi.org/10.1016/j.patcog.2019.106982