A robust supervised subspace learning approach for output-relevant prediction and detection against outliers. (October 2021)

• Record Type: Journal Article
• Authors: Li, Wenqing; Wang, Yue
• Is Part Of: Journal of process control. Volume 106(2021)
• Page Start: 184
• Page End: 194
• Publication Date: 2021-10
• Subjects: Output-related detection -- Robust supervised subspace learning (RSSL) -- Subspace decomposition -- Matrix factorization -- Industrial system; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2021.09.007
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract: This paper proposes a novel robust supervised subspace learning (RSSL) method for output-relevant prediction and detection against outliers. RSSL learns the robust subspaces by optimizing a joint problem over both the prediction of output and the reconstruction of input. To this end, the learned subspaces/data representations are informative, i.e., they are encapsulated with the critic information related to both the input and output, and thus can benefit the following tasks of output-related modeling and detection. Besides, we separate sparse items from the raw measurements to suppress the effects of outliers. An efficient optimization algorithm is designed to solve the optimization problem of RSSL. We further conduct post orthogonal decomposition upon the subspaces provided by RSSL so that the trimmed subspaces are more suitable for output-related detection. The efficacy of the proposed method is extensively verified by synthesis data and benchmark data. Highlights: We learn robust latent subspaces by a joint optimization of input and output, where the negative effects of outliers are suppressed by separating sparse items from the original data. We develop an efficient algorithm to solve the proposed formulation of RSSL. The theoretical and empirical analysis demonstrate the effectiveness of the designed optimization algorithm. The proposed RSSL is competent to the tasks of output prediction and output-related detection, where, for the latter task, post-decomposition is performed to further trim the subspaces generated by RSSL.