Tensorized Multi-View Multi-Label Classification via Laplace Tensor Rank

Qiyu Zhong, Yi Shan, Haobo Wang, Zhen Yang, Gengyu Lyu

Abstract: In multi-view multi-label classification (MVML), each object has multiple heterogeneous views and is annotated with multiple labels. The key to deal with such problem lies in how to capture cross-view consistent correlations while excavate multi-label semantic relationships. Existing MVML methods usually employ two independent components to address them separately, and ignores their potential interaction relationships. To address this issue, we propose a novel Tensorized MVML method named TMvML, which formulates an MVML tensor classifier to excavate comprehensive cross-view feature correlations while characterize complete multi-label semantic relationships. Specifically, we first reconstruct the MVML mapping matrices as an MVML tensor classifier. Then, we rotate the tensor classifier and introduce a low-rank tensor constraint to ensure view-level feature consistency and label-level semantic co-occurrence simultaneously. To better characterize the low-rank tensor structure, we design a new Laplace Tensor Rank (LTR), which serves as a tighter surrogate of tensor rank to capture high-order fiber correlations within the tensor space. By conducting the above operations, our method can easily address the two key challenges in MVML via a concise LTR tensor classifier and achieve the extraction of both cross-view consistent correlations and multi-label semantic relationships simultaneously. Extensive experiments demonstrate that TMvML significantly outperforms state-of-the-art methods.

Lay Summary: Modern data, like news webpage data, often comes in multiple forms—such as text, images, and videos—and can belong to multiple categories at once. Existing methods struggle to analyze these complex relationships effectively because they treat different data types and labels separately. Our work introduces a new approach called TMvML, which combines multi-view and multi-label learning into a unified framework using tensor-based techniques. Whatsmore, we develop a novel Laplace Tensor Rank (LTR) that better identifies meaningful patterns while filtering out noise. Experiments on real-world datasets show that TMvML achieves higher accuracy than current state-of-the-art methods, making it a practical solution for tasks like music or biological classification where multi-modal data and multiple labels are common. By capturing interactions between different data types and their labels in a single model, TMvML provides a more efficient and interpretable way to handle complex classification problems.

Primary Area: General Machine Learning->Supervised Learning

Keywords: Multi-view learning, Multi-label Learning

Submission Number: 1263