Multi-label feature selection via robust flexible sparse regularization

• Published in: Pattern recognition Vol. 134; p. 109074
• Main Authors: Li, Yonghao, Hu, Liang, Gao, Wanfu
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.02.2023
• Subjects: Classification; Feature selection; Multi-label learning; Sparse regularization; Multi-label learning; Feature selection; Sparse regularization; Classification
• ISSN: 0031-3203; 1873-5142
• DOI: 10.1016/j.patcog.2022.109074

•A regularization norm named robust flexible sparse regularization (RFSR) is designed.•The proposed norm overcomes the limitations of existing norms.•RFSR is introduced into the proposed framework with the global solution.•An efficient optimization method to solve the proposed method is designed. Multi-label feature selection is an efficient technique to deal with the high dimensional multi-label data by selecting the optimal feature subset. Existing researches have demonstrated that l1-norm and l2,1-norm are promising roles for multi-label feature selection. However, two important issues are ignored when existing l1-norm and l2,1-norm based methods select discriminative features for multi-label data. First, l1-norm can enforce sparsity on each feature across all instances while numerous selected features lack discrimination due to the generated zero weight values. Second, l2,1-norm not only neglects label-specific features but also ignores the redundancy among features. To this end, we design a Robust Flexible Sparse Regularization norm (RFSR), furthermore, proposing a global optimization framework named Robust Flexible Sparse regularized multi-label Feature Selection (RFSFS) based on RFSR. Finally, an efficient alternating multipliers based optimization scheme is developed to iteratively optimize RFSFS. Empirical studies on fifteen benchmark multi-label data sets demonstrate the effectiveness and efficiency of RFSFS.