Weight matrix sharing for multi-label learning

doi:10.1016/j.patcog.2022.109156

Weight matrix sharing for multi-label learning

2023-04-01

21

PRE

AI

DOI：10.1016/j.patcog.2022.109156

Original

Original request for help

Save

Abstract

En 中文

Multi-label learning on real-world data is a challenging task due to sparse labels, missing labels, and sparse structures. Some existing approaches are effective in addressing the former two issues. In this pa-per, we propose a shared weight matrix with low-rank and sparse regularization for multi-label learning (2SML) algorithm to address the issues simultaneously. First, two explicit correlation matrices are con-structed from the feature matrix and label matrix. Second, we select informative labels by instance rep-resentativeness to learn implicit correlations. Third, a feature manifold and label manifold are employed to guide the shared weight learning process. Extensive experiments are undertaken on multiple bench-mark datasets with and without missing labels. The results show that the proposed method outperforms the state-of-the-art methods.(c) 2022 Elsevier Ltd. All rights reserved.

Keywords:

Low-rank

Missing labels

Multi -label learning

Shared weight

Sparse

Weight matrix sharing for multi-label learning

Abstract

Journal

Organization