The study focuses on unsupervised multi-view representation learning for mining multi-view data. It identifies issues with existing methods, such as the inability to fully explore multi-view data and the lack of interpretability in kernel or neural network methods. The authors propose a new multi-view fuzzy representation learning method based on the Takagi-Sugeno-Kang (TSK) fuzzy system. Their method transforms multi-view data into a high-dimensional fuzzy feature space and introduces a regularization method to mine consistency information between views. The proposed method was tested on benchmark multi-view datasets to validate its effectiveness.
Publication date: Not provided
Project Page: Not provided
Paper: https://arxiv.org/pdf/2309.11473
