TY - GEN
T1 - A Non-iterative Partial Least Squares Algorithm for Supervised Learning with Collinear Data
AU - QIN, S. Joe
N1 - Financial support for this work from the Natural Science Foundation of China grant (U20A201398), Big data-driven abnormal situation intelligent diagnosis and self-healing control for process industries, and City University of Hong Kong Project (9380123), SGP: Bridging between Systems Theory and Dynamic Data Learning towards Industrial Intelligence and Industry 4.0, is gratefully acknowledged.
PY - 2021/12
Y1 - 2021/12
N2 - Partial least squares (PLS) has gained popularity in many domains such as industrial internet of things, bio-informatics, and econometrics due to its ability to deal with limited data, collinearity, and relevance to supervised machine learning. PLS has also been applied to system identification and subspace identification where the model is high-order, high-dimension, or collinear due to the lack of rich excitation. How-ever, all PLS algorithms to date are iterative in calculating the sequence of latent variables, unlike other related methods such as principal component regression. The iterative PLS estimation has made it difficult to perform statistical analysis. In this paper, we propose a novel non-iterative PLS algorithm based on the Krylov sequence used in PLS algorithms. Only a singular value decomposition is needed to obtain an equivalent PLS model for multiple PLS latent factors. The non-iterative PLS algorithm extracts the same latent space as the conventional PLS, which is demonstrated with a couple of industrial application cases.
AB - Partial least squares (PLS) has gained popularity in many domains such as industrial internet of things, bio-informatics, and econometrics due to its ability to deal with limited data, collinearity, and relevance to supervised machine learning. PLS has also been applied to system identification and subspace identification where the model is high-order, high-dimension, or collinear due to the lack of rich excitation. How-ever, all PLS algorithms to date are iterative in calculating the sequence of latent variables, unlike other related methods such as principal component regression. The iterative PLS estimation has made it difficult to perform statistical analysis. In this paper, we propose a novel non-iterative PLS algorithm based on the Krylov sequence used in PLS algorithms. Only a singular value decomposition is needed to obtain an equivalent PLS model for multiple PLS latent factors. The non-iterative PLS algorithm extracts the same latent space as the conventional PLS, which is demonstrated with a couple of industrial application cases.
UR - http://www.scopus.com/inward/record.url?scp=85126041417&partnerID=8YFLogxK
U2 - 10.1109/CDC45484.2021.9683510
DO - 10.1109/CDC45484.2021.9683510
M3 - Conference paper (refereed)
AN - SCOPUS:85126041417
SN - 9781665436601
T3 - Proceedings of the IEEE Conference on Decision and Control
SP - 3683
EP - 3688
BT - 2021 60th IEEE Conference on Decision and Control (CDC)
PB - Institute of Electrical and Electronics Engineers
T2 - 60th IEEE Conference on Decision and Control (CDC 2021)
Y2 - 13 December 2021 through 17 December 2021
ER -