My research focuses on pattern recognition, principal component analysis, multi-view clustering, and partial multi-view clustering.

Generative Adversarial Network (GAN) Based Incomplete Multi-View Clustering

To improve the performance of the incomplete multi-view clustering, we propose to GAN-based incomplete multi-viewclustering. We also apply the proposed method into multi-view clustering for better performance.

Related Publications

[1] Qianqian Wang, Zengming Ding, Zhiqiang Tao, Quanxue Gao and Yun Fu, Partial Multi-view Clustering via Consistent GAN, IEEE ICDM, 2018: 1290-1295.

[2] Zhaoyang Li, Qianqian Wang, Zhiqiang Tao, Quanxue Gao, and Zhaohua Yang. Deep Adversarial Multi-view Clustering Network, IJCAI, 2019:2952-2958.

Canonical Correlation Analysis (CCA) Based Deep Cross-Modal Clustering

Existing deep cross-modal clustering cannot well exploit the consensus structure embedded in multi-modal data, especially when data of each modal shows great variety. To overcome this problem, we propose to employ CCA to obtain the consensus structure of them and propose CCA-based deep cross-domain clustering method. Additionally, we further extend this work to handle the incomplete cross-modal data processing and design a CCA-based deep incomplete cross-modal data clustering method.

Related Publications

[1] Quanxue Gao, Huanhuan Lian, Qianqian Wang, Gan Sun: Cross-Modal Subspace Clustering via Deep Canonical Correlation Analysis, AAAI, 2020.