Auto-weighted multi-view clustering via deep matrix decomposition

作者：

Highlights：

• Anovel deep multi-view learning model is proposed by uncovering the hierarchical semantics of the input data in a layer-wise way.

• The instances from the same class but from different views are forced to be closer layer by layer in the low-dimensional space, which is beneficial for the subsequent learning task.

• To automatically determine the weights of different views, we introduce the auto-weighting scheme into the deep multi-view clustering algorithm.

• To solve the optimization problem of our model, an efficient iterative updating algorithm is proposed with a theoretical guarantee of its convergence.

摘要

•Anovel deep multi-view learning model is proposed by uncovering the hierarchical semantics of the input data in a layer-wise way.•The instances from the same class but from different views are forced to be closer layer by layer in the low-dimensional space, which is beneficial for the subsequent learning task.•To automatically determine the weights of different views, we introduce the auto-weighting scheme into the deep multi-view clustering algorithm.•To solve the optimization problem of our model, an efficient iterative updating algorithm is proposed with a theoretical guarantee of its convergence.

论文关键词：Multi-view learning,Deep matrix decomposition,Clustering,Optimization algorithm

论文评审过程：Received 16 February 2019, Revised 8 August 2019, Accepted 17 August 2019, Available online 28 August 2019, Version of Record 30 August 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107015